Transcript

SYNTHETIC AND ANAL YTICAL STUDIES

OF

810MIMETIC METAL COMPLEXES

THESIS

Submitted in fulfilment of the requirements for the Degree of

DOCTOR OF PHILOSOPHY

of Rhodes University

by

KEVIN WAYNE WELLINGTON

Department of Chemistry

Rhodes University

June 1999

TABLE OF CONTENTS

page

1 INTRODUCTION 1

1.1 METALLOENZYMES 1

1 . 1 .1 Background 1

1.1.2 The structure of metalloenzymes 2

1.1.3 Functions of metalloenzymes 3

1.2 TYROSINASE 5

1.2.1 Background 5

1.2.2 The active sites of tyrosinase and haemocyanin 8

1.2.3 Studies of the active site of haemocyanin 9

1.2.3. 1 Early studies 9

1.2.3.2 More recent studies 10

1.2.4 Similarities between tyrosinase and haemocyanin 11

1.2.5 A possible mechanism of oxidation by tyrosinase 13

1.2.6 Applications of the enzyme tyrosinase 15

1.3 THE BIOMIMETIC APPROACH TO TYROSINASE ANALOGUES 16

1.3.1 The selection of ligands for dinuclear copper complexes ' 18

1.3.2 Coordination considerations in the reaction of copper

complexes with dioxygen 20

1.3.3 The mode of binding of dioxygen in oxyhaemocyanin and

oxytyrosinase 22

1.3.4 Synthetic biomimetic models of haemocyanin and

tyrosinase 24

1.4 THE IMPORTANCE OF METAL-DIOXYGEN BONDING 37

1.5 OBJECTIVES OF THE PRESENT INVESTIGATlON 41

II

2 DISCUSSION 42

2.1 LIGAND DESIGN AND RATIONALE 42

2.2 LIGAND SYNTHESES 43

2.2.1 Biphenyl ligands 44

2.2.2 1,1 O-Phenanthroline ligands 54

2.2.3 Schiff base ligands prepared from diketones 59

2.2.4 The Baylis-Hillman approach to ligand syntheses 64

2.2.5 Macrocycle syntheses 70

2.2.6 Dendrimer synthesis 73

2.3 COMPLEXATION AND COMPUTER MODELLING STUDIES 78

2.3.1 Complexation Studies 78

2.3.2 Copper Complexes 80

2.3.3 Computer Modelling Studies 90

2.3.4 Other Transition Metal Complexes 101

2.3.4.1 Cobalt Complexes 101

2.3.4.2 Nickel Complexes 114

2.3.4.3 Platinum Complexes 127

2.4 ELECTROCHEMICAL STUDIES: CYCLIC VOL TAMMETRY~ 136

2.4.1 Copper Complexes 1137

2.4.2 Cobalt Complexes 142

2.4.3 Nickel Complexes 145

2.5 BIOMIMETIC EVALUATION OF THE CATALYTIC ACTIVITY OF

COPPER COMPLEXES 150

2.6 CONCLUSIONS 155

iii

3 EXPERIMENTAL 157

3.1 GENERAL 157

3.2 PROCEDURES FOR LIGAND SYNTHESES 158

3.2.1 Biphenyl ligands 158

3.2.2 1,1 O-Phenanthroline ligands 167

3.2.3 Schiff base ligands prepClred from diketones and

2-(2-aminoethyl)pyridine 172

3.2.4 The Baylis-Hillman approach to ligand synthesis 177

3.2.5 Macrocycle syntheses 181

3.2.6 Dendrimer synthesis 185

3.3 SYNTHETIC PROCEDURES FOR THE PREPARATION OF

COMPLEXES 185

3.3.1 Copper Complexes 185

3.3.2 Cobalt Complexes 190

3.3.3 Nickel Complexes 193

3.3.5 Platinum Complexes 195

3.4 PROCEDURE FOR THE COMPUTER MODELLING OF

COMPLEXES 198 !

3.5 ELECTROCHEMICAL STUDIES: CYCLIC VOL TAMMETRY 199

3.6 METHOD FOR THE EVALUATION OF THE CATALYTIC

ACTIVITY OF COPPER COMPLEXES 205

4 REFERENCES 206

5 APPENDIX: CRYSTAL STRUCTURE DATA 220

IV

ACKNOWLEDGEMENTS

I am grateful to my supervisor Prof. Perry Kaye for granting me the opportunity to

work with him and also for the advice, support and encouragement over the years.

Also to Prof. Nyokong, my co-supervisor who has, over the years, been a source

of inspiration and encouragement, for her support and advice with the

electrochemistry. I would also like to thank Dr. Cheryl Sacht for informative

discussions concerning coordination chemistry. Special thanks go to Dr. Gary

Watkins, also for informative discussions, but most of all for his generous support

and guidance with the coordination chemistry section despite not being one of my

supervisors.

I am also grateful to my colleagues, who over the years, have made my stay in the

department a pleasant one, for their support and encouragement. I want to thank,

in particular, Andre Daubinet who has always been very willing to assist with the

molecular modelling programme Cerius2.

\ I am very grateful to Mintek and, also to the FRO and DAAD for their financial

J support. I want, in particular, to thank Mr. Bob Tait, Dr. Kathy Sole and Dr. Brian

Green at Mintek.

I also want to thank the technical staff (Mr Aubrey Sonne man and Mr. Andre

Adriaan), the secretaries (Mrs. Margie Kent and Mrs. Bernita Tarr), and the

storeman (Mr. Johan Buys) for their support, and Ms Leanne Cook (University of

the Witwatersrand) for X-ray structure determinations.

Most of all, I would like to thank my parents for their support and encouragement

over the years. I dedicate this thesis to my parents in appreciation of their support

and encouragement.

v

ABSTRACT

Several series of novel diamido, diamino and diimino ligands containing different

spacers and heterocyclic donors have been synthesised. The spacers include the

flexible biphenyl, the rigid 1,1 O-phenanthroline and various acyclic moieties, while

the heterocyclic donors comprise pyridine, imidazole or benzimidazole groups.

These ligands have been designed to complex copper and act as biomimetic

models of the active site of the enzyme, tyrosinase, and their complexes with

copper, cobalt, nickel and platinum have been analysed using microanalytical, IR,

UV-Visible and cyclic voltammetric techniques. Attempted reduction of the

biphenyl-based diimino ligands resulted in an unexpected intramolecular

cyclisation affording azepine derivatives, the structures of which were elucidated

with the aid of single crystal X-ray analysis of cobalt and nickel complexes.

Computer modelling methods have been used to explore the conformational

options of the copper complexes, and to assess the accessibility of the dinuclear

copper site to substrate molecules. Computer modelling has also been used, in

conjunction with the available analytical data, to visualise the possible structures

of selected ligands and complexes. I

The copper complexes, although predominantly polymeric, were evaluated as

biomimetic catalysts using 3,5-di-t-butylphenol and 3,5-di-t-butylcatechol as

substrates. Some of the complexes clearly displayed biomimetic .. potential,

exhibiting both phenolase and catecholase activity.

vi

ABBREVIA TIONS

COl carbonyldiimidazole

CH2CI2 dichloromethane

CHCI3 chloroform

CT charge transfer

CV cyclic voltammetry

OTSC 3,5-di-tert-butylcatechol

OTSP 3,5-di-tert-butylphenol

OMF N, N-dimethylformamide

OMSO dimethylsulfoxide

Et3N triethylamine

Et20 diethyl ether

EPR electron paramagnetic resonance

EXAFS extended X-ray absorption fine structure

HCSO hexachlorobutadiene

IR infrared

MeCN acetonitrile

NMR nuclear magnetic resonance .I

THF tetrahydrofuran

UV ultraviolet

1

INTRODUCTION

1.1. METALLOENZYMES

1.1.1. Background

Metal ions are often found as constituents of proteins and enzymes. In fact,

approximately one third of enzymes known require metal ions for activitY,1 and it

is apparent that nature has utilized the special properties of metal ions to perform

a wide variety of specific functions which may be crucial in life processes. In

Australia, for example, the deficiency of copper in certain areas adversely affected

the nervous system of sheep and caused ,pnaemia and wool deterioration.1

Metal ions in proteins and enzymes can be roughly divided into two classes, viz.,

chemical and structural metals. Metals which enter directly into biological reactions

in a chemical manner are referred to as chemical metals, e.g. iron(ll) in its role in

the oxidation-reduction reactions catalysed by peroxidases and ferrodoxins or the

bonding of oxygen by haemoglobin. Those metals that stabilise ~the protein for

biological function (e.g. calcium(lI) in thermolysin) or indirectly promote c9talysis

by inducing a required orientation of th.e substrate or catalytic group in the protein

(e.g. magnesium(lI) in phosphoglucomutase) are referred to as structural metals.1

It has also been observed2 that metal ions in proteins and enzymes are able to

perform in two ways: i) as an integral part of the protein, removable only by

extreme chemical attack and exhibiting high metal-ion specificity in function, or ii)

loosely bound to the substrate or enzyme, readily dialysable and exhibiting low

metal-ion specificity in catalytic function.

Introduction 2

1.1.2. The Structure of Metalloenzymes

Metalloenzymes contain metal ions that are tightly bound and always isolated with

the protein. The structure of the enzyme is destroyed upon removal of the metal

ions. The metals ions are mainly responsible for electron transfer. Alcohol

dehydrogenase (containing Zn2+),3 arginase (containing Mn2+),4 ferredoxin

(containing Fe2+),5 haemocyanin (containing CU2+)5 and urease (containing Ni2+)6

are all metalloenzymes. In the case of the proteins, haemoglobin and the

cytochromes, however, the metal (Fe2+ or Fe3+) is part of the haeme prosthetic

group (non-protein). Metal ions frequently occur within the active site of the

enzyme and resemble protons in that they are electrophiles, which are able to

accept an electron pair for the formation of a coordinate-covalent bond with anionic

or neutral ligands to yield complexes of various geometries, e.g., linear, square

planar, tetrahedral or octahedral. 7

The ability of free amino and carboxyl groups in proteins to bind to metals allows

for the establishment of an active conformation, as in the case of carboxypeptidase

A 8 This enzyme, which promotes hydrolysis, has Zn2+ in its active site. When the

ester substrate fits into the hydrophobic pocket of the enzyme, the,zinc stabilises

the enzyme-substrate complex by coordinating to the developing oxygen allion as

the water attacks the carbonyl carbon of the ester substrate. In carbonic

anhydrase, zinc binds to water making it sufficiently acidic to permit the loss of a

proton and formation of a nucleophilic-hydroxyl group.

It is known that the biological role of metals in metalloenzymes is highly specific,

and that the same metal ion in a different protein is able to perform different

functions. The range of functions promoted, however, is largely dictated by the

protein environment.

Introduction 3

1.1.3. Functions of Metalloenzymes

Metalloenzymes function in many essential physiological processes. They effect

a variety of important chemical transformations, often involving small molecule

substrates or products such as molecular oxygen, hydrogen, nitrogen and water.

The transformations occur with exceptional substrate regio- and stereoselectivity

under mild conditions. Various metal clusters or ions have been observed at the

active sites, and living systems have utilised rudimentary recurring, structures that

have been modified or adapted for particular purposes. The role of

metalloenzymes is to perform specific catalytic functions. Dioxygen metabolism is

vital to aerobic organisms, mainly as a primary energy source having a

thermodynamically favourable reduction to water [standard potential, EO = 0.82 V

(1 atm 02, pH = 7)], and metalioenzymes have been used for the insertion of

dioxygen into biological substrates.9 This insertion occurs through mono- or

dioxygenation processes. The functions of various metalioproteins and

metalloenzymes are summarised in Table 1.10

Metalloenzymes may be controlled by changing the pH, thus disrupting the flow of

electrons which the enzyme would normally regulate. Metalloenzym~s can also be

inhibited by using transition state analogs which mimic the structure! of the

transition state in the reaction of the enzyme with a particular substrate.

Introduction 4

Table 1: Functions of various metalloenzymes and metalloproteins. 10

Metal Protein Source Function

Mo nitrogenase Cl.pasteurium electron transfer

nitrate reductase N. crassa electron transfer

Co methionine micro-organisms methionine biosynthesis, transfer

synthetase of a methyl group

Fe haemerythrin blood cells of marine worms oxygen storage and/or transport

haemoglobin mammalian erythrocytes oxygen storage and/or transport

myoglobin mammalian muscle oxygen storage and/or transport

cytochrome band c animal and plant tissue, liver, electron transfer in mitochondrial

catalase eryth rocytes respiration

peroxidase plant roots ~> oxidation of H20 2 to 02

cytochrome oxidase yeast and animal tissue reduction of HP2 to HP

Fe, Cu haemocyanin blood of molluscs electron transfer in O2 reduction

Cu tyrosinase mushrooms oxygen transport

dopamine-~- adrenal medulla mixed function oxidase

hydroxylase

\

Cu, Zn erythrocuprein erythrocytes mixed function oxidase .I

Zn carboxypeptidase pancreas superoxide dismutation

carbonic anhydrase erythrocytes protein hydrolysis

alcohol liver acid-base control, CO2 hydration-

dehydrogenase dehydration metabolism and

oxidation of alcohols

Zn, Ca thermolysin B. thermoproteolyticus Zn, protein hydrolysis; Ca,

thermal stability

Mn pyruvate carboxylase liver formation of pyruvic acid from

Acetyl CoA

Mn, Ca conconavalin A jack bean cell mitogen

- <

Ca a-amylase B.subtilis, saliva carbohydrate hydrolysis

-

Introduction 5

1.2 TYROSINASE

1.2.1 Background

Tyrosinase is a metalloenzyme which was discovered in 1895, when a black pigment

was observed in the mushroom, Russulla nigricans. 11 The enzyme was named

tyrosinase in reference to the substrate tyrosine, oxidation which affords a black

pigment. As the structures of most tyrosina~es and their modes of action have not

been fully elucidated, tyrosinase remains the subject of wide-ranging research by

many biologists and chemists. Tyrosinase is found widely in microorganisms, plants

and animals. In microorganisms and animals, tyrosinase catalyses the initial step in

the biochemical formation of the highly coloured pigment, melanin, from tyrosine. 11

In plants, the physiological substrates comprise a variety of phenolics, which are

oxidised by tyrosinase in the browning reaction observed when apples, bananas,

mushrooms etc. are injured and exposed to dioxygen. Although the function of this

reaction has not been established, it is believed that it may offer protection of the

wound from pathogens or insects. It has also been suggested that tyrosinase may be

involved in wound healing and, possibly, sclerotisation of the cuticle in insects. 11

OH OH II 0 " ¢ ~,2H+ qOH H2O ¢r0+ ~ I . ~ 2H+ (eq. 1)

tyrosinase ~ tyrosinase

R R R R=Me

tyrosine DOPA DOPAquinone

Tyrosinase catalyses both the o-hydroxylation of monophenols (cresolase activity)

and the two-electron oxidation of the o-diphenol (catecholase activity) to the 0-

quinone using molecular O2 .11 The substrate supplies the two electrons required for

the reduction of the second oxygen atom to H20 and, thus, functions as an internal

monooxygenase. The hydroxylation reaction (I) is faster than the oxidation reaction

Introduction 6

(II) (k(,) = 103 S-1, k(ll) = 107 S-1; eqn.1).12 The rate determining step is believed to be

the hydroxylation of the substrate to DOPA (dihydroxyphenylalanine). The pigment,

melanin, is obtained via a series of non-enzymatic polymerisation reactions of the

DOPAquinone product. Other common names for tyrosinase are phenolase and

polyphenol oxidase.

Table 2: Properties of some tyrosinases which have been well characterised. 13

* * Absorption nm Circular dichroism I Amino

Source No. of MW/subunits (x 10-3 M-1cm-1)

nm (x 10-3 M-1 cm2 acid

subunits (KDa) dmol-1 ) sequence

Streptomyces 1 30.9 345 (17.4) 345 (-32.5) yes

g/augescens 640 (1.5) 470 (2.1) "" ...

575 (-1.7)

740 (5.0)

Neurospora 1 46 345 (18) 345 (-27) yes

crassa 425 (0.5) 520 (0.6)

600 (1.0) 600 (1.0)

750 (3)

Agaricus 2 13.4 345 (18) 353 no

bisporus 2 43 600 (1.2)

(Mushroom) \

*Absorption and Circular Dichroism of OXY form

Tyrosinases isolated from the eubacteria Streptomyces glaucescens and the fungi

Neurospora crassa and Agaricus bisporus are the best characterised (See Table 2).

Although tyrosinase is the most widely studied multicopper oxygenase, there are

many other important copper proteins (see Table 3) which are involved in the

reactions of dioxygen or its derivatives.

Introduction 7

Table 3: Other important copper proteins 14,15

Protein Enzyme Source Function

"Blue" electron carriers Azurin Algae, green leaves and Electron transfer (Photosynthesis)

Plastocyanin other plants

Stellacyanin

Umecyanin

"Blue" oxidases Laccase Tree, Fungal Oxidation of phenols and diamines

Ceruloplasmin Human, animal serum Weak oxidase activity

Fe and Cu transport

Ascorbate Oxidase Plants Oxidation of L-Ascorbic acid

Oxygen carrier Haemocyanin Molluscs and arthropods Oxygen transport

Copper monooxygenases Phenol o-monooxygenase Animal skin, plants, tyrosine oxidation to melanin

insects, melanoma

Dopamine beta-hydroxylase Adrenals Converts dopamine to

norepinephrine

Introduction 8

1.2.2 The active sites of tyrosinase and haemocyanin

There are inherent difficulties in purifying multi-subunit enzymes, and the elucidation

of the protein structure of tyrosinases has been hampered by such multiplicity. As a

result, there is no complete structure presently available for any tyrosinase enzyme.

However, some understanding of the nature and function of the active site can be

obtained by correlations with haemocyanins,16 for which crystal structures of both the

deoxy17 and oxi8 forms of the active sites have been obtained. Based on

spectroscopic properties (mainly EPR), the copper proteins have been divided into

three main groups (see Table 4).

Table 4: The groups of copper proteins and their characteristics. 19

Group Characteristics Examples

Type I Mononuclear copper having a trigonal basal plane with an N28 ligand donor set Plastocyanin

(8 denotes thiolate sulfur from cysteine); exhibits unusual spectroscopic Azurin

properties. viz.,

(1) a strong absorption band at ca. 600nm (an intense blue colour),

(2) a small All value «70 G), and

(3) a high reduction potential (generally >250 mV).

Type II Mononuclear copper site exhibiting a normal EPR spectrum. ) . I

Phenylalanme hydroxy ase,

Three sub-groups have been identified, viz., IIA, liB and IIC. galactose oxidi!se,

superoxide dismutase.

Type IIA The ligands comprise ordinary protein residues, such as histidine imidazole,

cysteine thiolate and water (or hydroxide).

Type liB The ligand donors include unusual protein side chains.

Type IIC The copper is bridged to another metal ion, forming a hetero-dinuclear metal

site.

Type III EPR silent dinuclear site which binds dioxygen as peroxide and exhibits Tyrosinase and

unusual physicochemical characteristics, viz., haemocyanin

(1) diamagnetism,

(2) two characteristic absorption bands at 350 and 580 nm, and

(3) a low 0-0 vibratiQ!) stretching frequency (ca. 750 cm").

Introduction 9

Tyrosinase and haemocyanin have been classified as Type III copper proteins. Such

proteins are characterised by a dinuclear copper active site, the copper atoms being

anti-ferromagnetically coupled in the oxidised state and, therefore, EPR silent. 20

1.2.3 Studies of the active site of haemocyanin

1.2.3.1 Early studies

Haemocyanins function as oxygen carriers in the haemolymph of molluscs and

arthropods. They have a dinuclear copper active site which bind oxygen, as a

peroxide, to give oxyhaemocyanins. Volbeda and Hol21 were the first to determine the

three-dimensional structure of the deoxygenated form of haemocyanins from

Panuiirus interruptus, and found that the two Cu(l) ions of the active site are

embedded in a protein matrix with three histidine residues coordinating each copper

ion.

Table 5: Data for haemocyanin derivatives with different active site arrangements of the copper ions. 29

Derivative Name Properties

Cu(I) .. Cu(l) deoxy Coupled, colourless ,

CU(II).02·Cu(II) oxy Coupled, absorbs at 330 and 650 nm J

Cu(II) .. Cu(l) half-met EPR-active

Cu(II) .. Cu(II) met Coupled, EPR-silent

Cu(II) .. - met-apo Paramagnetic

Cu(II) Cu(II) dimer Paramagnetic, not coupled

Earlier, chemical and spectroscopic evidence suggested that each cuprous ion in

deoxyhaemocyanin was coordinated to two or three imidazole ligands from

histidine. 22-28 The spectral.!eatures of various forms of haemocyanin, generated with

different active site arrangements, were investigated by-Lang and Holde29 '(Table 5).

The charge transfer spectra and magnetic properties of the oxy derivatives indicated

Introduction 10

that oxygen was bound as an end-to-end peroxide (02-) bridge across the two copper

atoms. It was proposed that the oxygen bridge was positioned in the equatorial plane

between the two copper atoms, as in Figure 1.

Figure 1

~N HiS" 0 His H" / 'c / IS-CU u, His

/ \ / , His 0-0 His

From an X-ray crystallographic study of Panulirus interruptus haemocyanin, the Cu(I)­

Cu(l) distance in the deoxy form was found to be 3.8 A.30 Upon reaction of the

deoxyhaemocyanin with O2, significant coordination takes place giving rise to

tetragonally coordinated copper(II) ions separated by 3.6 A. These copper(II) ions are

bridged by 0/-and an endogenous oxygen-containing group.31

1.2.3.2 More recent studies

Crystals of haemocyanin and oxyhaemocyanin from Limulus po/iphemus have also

been analysed.32,33 High resolution X-ray crystal data for the oxyhaemocyanin

indicated that each copper atom is tightly coordinated in a square\planar geometry

to both atoms of the oxygen molecule and to the nitrogen atoms of the two closer

histidine ligands (Figure 2). Each copper atom is also weakly coordinated to the third

histidine ligand which is in an axial position. Molecular oxygen lies in a side-on (IJ­

rtf() configuration with respect to the Cu-Cu axis. The two copper, the two oxygen,

and the four nitrogen atoms lie approximately in the same plane. The two copper

atoms are 3.54 A apart while the two oxygen atoms are 1.41 A apart. The relative

positions of the copper atoms and histidine residues in this structure are very similar

to those found by Volbeda and HoI. 21

Figure 2

Introduction

His \ N

H" I IS-N : 0 N-His "", //' / /Cu, /C;;u,

His-N 0: N-His I

N 'His

11

The data obtained for the deoxygenated form of the haemocyanin obtained from L.

po/iphemus, however, are rather different from those reported for the P. interruptus

protein. Each copper atom exhibits trigonal geometry and is coordinated to three

histidines, while the two copper atoms are 4.6 A apart - a distance greater than the

value found for the deoxygenated P. interruptus haemocyanin (3.8 A).21

The structural difference between these two deoxy forms of haemocyanin has been

explained in terms of a two-state model. The haemocyanin hexamer may adopt

conformational states of high (R-state) or low (T-state) oxygen affinity, either of which

may be oxygenated or deoxygenated. It has been postulated that the deoxygenated

L poliphemus haemocyanin is in the T-state, while that of P. interruptus haemocyanin

assumes the deoxygenated R-state. 32

J

1.2.4 Similarities between tyrosinase and haemocyanin

Chemical and spectroscopic studies indicate that tyrosinase has a coupled dinuclear

copper active site very similar to that of haemocyanin. Studies performed on a series

of derivatives of mollusc and arthropod haemocyanins34-41 and fungal tyrosinase, 42-46

have shown that the active sites of these proteins are remarkably similar and include

examples of the deoxy [Cu(l) Cu(l)], oxy [Cu(lI) O2 Cu(II)], mixed-valent half-met [Cu(I)­

Cu(II)], met [Cu(II)-Cu(II)]and the dimer [Cu(II)' "Cu(II)] arrangements (seeTable 5).

- The met derivative is considered to be the resting form of tyrosinase but,

Introduction 12

unfortunately, there is currently no crystal structure available for the corresponding

met haemocyanin. The EPR silence of the met form is due to antiferromagnetic

coupling between the two copper(lI) ions, which requires a super-exchange pathway

involving a bridging ligand. 40 The crystal structure of P. interruptus haemocyanin

reveals that it is not possible for the protein residues, in the vicinity of the copper site,

to provide such bridging; in this case, the bridging ligand is presumed to be a water

hydroxyl group.21

It is believed that a similar situation exists at the active site in met tyrosinase. It

appears that only one water-derived ligand is bound terminally to the two copper(lI)

ions as the met tyrosinase Cu-Cu distance of 3.4 A (as determined by EXAFS

analysis) is too large for two, single atom hydroxide bridges. 16 Oxytyrosinase is

produced from met tyrosinase by the addition of peroxide, or by two electron

reduction to the deoxy state followed by the reversible binding of dioxygen.43,44,47 As

oxytyrosinase reacts with both the monophenol and the diphenol, 17,18 the geometric

and electronic structures are important in understanding the hydroxylation chemistry

of this enzyme. For oxyhaemocyanin, the following spectroscopic features are

observed:- (i) an intense absorption band at 350 nm (8 -20000 M-1 cm-1);36 (ii) a low

0-0 stretching frequency of ca. 750 cm-1; 19 and (iii) a Cu-Cu distance of 3.6A (from

EXAFS16 and X-ray crystallography48). The spectral features observed for oxtyrosinase

are essentially the same,16,43,44,46,47 and are characteristic of the side-on I-J-r{ r]2

peroxide bridging mode (see Figure 3).

Figure 3: The geometric arrangement of the side-on 1-J-r]2: r]2 peroxide bridging mode in oxyhaemoq~anin and oxytyrosinase.

Introduction 13

In view of the similarities in the spectroscopic features of these oxy states, it is

believed that oxytyrosinase has a similar dinuclear copper active site to that of

oxyhaemocyanin. These spectroscopic features are quite well understood. 21 -23 The

intense absorption band at 350 nm indicates that the side-on peroxide is a very

strong a-donor ligand to the two copper(lI) atoms, while the low 0-0 stretching

frequency (at ca. 750 nm) indicates a weak 0-0 bond. The weakness of the 0-0

bond derives from the ability of the peroxide to act as a n-acceptor ligand; the n-back­

bonding shifts electron density into the peroxide a* orbital, which is highly anti­

bonding with respect to the 0-0 bond, thus activating it for cleavage.

1.2.5 The proposed mechanism of oxidation by tyrosinase

A mechanism, which takes into account the: geometric and electronic characteristics

of oxyhaemocyanin, has been proposed to account for the phenolase and

catecholase activity of tyrosinase. 18 While earlier mechanisms have interpreted the

complex kinetics of tyrosinase in terms of allosteric effects and two binding sites, 44,48,49

the mechanism outlined in Figure 4 accounts for all the kinetic and inhibition patterns

observed for this enzyme. 18 In the first step of the phenolase cycle (A), a phenol binds

in an axial position to one of the copper atoms of the oxy site. A trigonal bipyramidal \

rearrangement towards the equatorial plane then occurs, which orients the artha-J

position af the phenol for hydroxylation by peroxide. The resulting o-diphenolate

(met-D) is then oxidised to the quinone releasing the deoxy site for a further oxidation

cycle. In the catecholase cycle (8), the a-catechol is oxidised to the quinone after

reacting with the oxy site. From kinetic studies it has been found that bulky

substituents on the phenol ring dramatically reduce the cresolase activity, but not the

catecholase activity.18 It has been deduced that monophenolic substrates require

axial-to-equatorial rearrangement for artha-hydroxylation, but this is not a necessity

for diphenolic substrates undergoing simple electron transfer. The bridged bidentate

_ coordination mode illustrated for the met-D form in Figure 4 is supported by the fact

8

met

Introduction

Q~ oxy-D

o 0 N" I Il,O,,1 II,N

,;Cu 1 Cu ...... N "o~ N

~7'\ o

N"I II ,0" ",N Cu' i 'Cil

N" "'O~"N Y OH

I A

Q HO OH

N, II ,0, II ,N .CU~i .. Cu'" NON

oxy

~ Q

deoxy

met-O

o 0

A - cresolase cycle B - catecholase cycle

Figure 4: A mechanism proposed by Solomon et al. for the cresolase and catecholase activity of tyrosinase, 50

J

14

that o-diphenol, but not the m- and p-diphenol (0- and p-diphenol have approximately

the same redox potential) is oxidised by tyrosinase,50 It has been noted that bidentate

coordination to one copper atom is also a possibility,50

Notwithstanding the similarities, a major difference between the active sites of

haemocyanin and tyrosinase is indicated by the fact that tyrosinase can react with

Introduction 15

substrates whereas haemocyanin cannot. An explanation for this difference IS

provided by sequence comparisons of tyrosinases with the structurally defined

haemocyanins, from which it is apparent that an additional domain in the larger

haemocyanin protein blocks access to the dinuclear copper site.

1.2.6 Applications of the enzyme tyrosinase.

Current interest in applications of the tyrosinase enzyme is reflected in publications

which have appeared in the recent literature. Atlow et al. 51 have reported use of the

enzyme in the removal of phenols from industrial, aqueous effluents. The phenols are

converted to the corresponding o-quinones, which then undergo non-enzymatic

polymerisation to form water-insoluble aggregates. Another study has been

concerned with the removal of aromatic amines from industrial waste-water. 52

Following treatment with the enzyme, a colour change from colourless to dark-brown

was observed, but no precipitation occurred; however, treatment of the coloured

compounds with a combination of tyrosinase and a catioRic polymer coagulant,

containing an amino group, resulted in the precipitation of the enzymatic reaction

products. For heterogeneous applications, tyrosinase has been immobilised via free

amino groups on cation exchange resins that can be used rep~atedly, 53 a 100%

removal of phenols being achieved after 2 h, with only marginal reduqtion in the

activity even after 10 repeat treatme[1ts.

The development of new biosensors is a rapidly growing research field and the

tyrosinase enzyme has been used as a biosensor for detecting several phenols and

o-diphenols. Studies of this application were first undertaken by Schiller and co­

workers,54.55 and involved immobilising the enzyme in a polyacrylamide gel. An

enzyme-spectrometric method was used initially but, subsequently, an enzyme­

potentiometric method was developed. Campanella et al. 56 have reported an enzyme-

- amperometric method in which tyrosinase may be immobilised in three different ways

Introduction 16

and coupled to an oxygen gas-diffusion electrode. This method has been used for the

determination of phenol in several environmental matrices, and evaluated as a

promising alternative to more conventional methods.

Introduction 17

1. 3 THE BIOMIMETIC APPROACH TO TYROSINASE ANALOGUES

Metalloenzymes effect a variety of important chemical transformations that often

involve small molecule substrates, such as molecular oxygen, hydrogen, nitrogen

and water. Metalloenzyme-catalysed reactions also function under very mild

conditions, and affect transformations with exceptional efficiency and substrate

regio- and stereoselectivity. As a result, extensive research has been undertaken

in this field to design synthetic processes that approach this efficiency and

selectivity and, thus, mimic the natural systems. A multidisciplinary bioinorganic

approach has been adopted, which encompasses a variety of disciplines such as

inorganic chemistry, synthetic organic chemistry, biochemistry and spectroscopy.

The basic objectives of the biomimetic approach, as applied to the modelling of

metalloenzymes, can be outlined as follows:-

(a) to develop a biomimetic model that will account for the spectroscopic

characteristics and function of the metalloenzyme in terms of its structure;

(b) to compare the observed metalloenzyme properties with those of

the biomimetic model; and

(c) to exploit the use of synthetic models which exhibit catalytic ~ctivity

comparable to the metalloenzyme. 57 J

Two principal strategies exist for the preparation of biomimetic models, viz., total

synthesis and metal template condensation (the self-assembly approach). The

total synthesis method allows for the incorporation of selected properties, such as

geometry, polarity, hydrophil/icity, hydrophobicity and steric characteristics; in this

method, the metal is added after the ligand has been synthesised. In the metal

template condensation method, on the other hand, the complex adopts an optimal

geometry dictated by the f'tature of the metal ion. It is also important to consider the

- - basic premise in bioinorganic chemistry, viz., that ttie chemistry of the metal-

binding site is dependent Qn the immediate ~nvironment of the metal ion. In many

metal/oproteins, the amino acid side chains provide the coordination environment

Introduction 18

and, at times, the coordination-sphere of the metal ion may be completed by a

prosthetic group, such as a porphyrin ring. 58 Once synthesised, the synthetic

model can be compared with the metalloprotein, with respect to spectral and

physicochemical properties, coordination geometry, catalytic activity, and the

mechanism of action. This may then provide information relating to the effects of

the environment on the active site and/or to the intrinsic properties of the active

site.

Besides providing a basis for postulating biological reaction pathways and possible

metal-complex intermediates, extensive modelling efforts may yield small­

molecule biomimetic catalysts capable of effecting transformations with practical

applications. Such applications include the following:- (i) the removal of

environmental pollutants, such as pho_sphate ester pesticides or nitrogen

compounds; (ii) the hydrolysis of peptides or nucleic acids in biotechnological

processes; or (iii) selective dioxygen-mediated oxidation in drug and chemical

transformations. 59

1.3.1 The selection of ligands for dinuclear copper complexes

\

To design biomimetic models that will mimic the metalloenzyme, it is essential that

" the structure of the model be as similar as possible to that of the metalloenzyme.

The ligands selected should be chemically similar to those in the enzyme and

should also have the correct number of donor atoms required for coordination to

the metal ions. For tyrosinase biomimetics, the ligands should be similar to the

histidyl-imidazole moieties. The sizes of the ligands are also significant as they

must be able to accommodate not only both copper ions, but also any additional

bridging group essential for reactivity. Thus, the ligands used must be sufficiently

flexible to adapt to different geometries, especially if there is a valence change, as

_ is the case in dioxygen binding. The geometriC and coordination requirements of

Cu(l) and Cu(lI) ions are CLuite different, as i~dicated in Table 2.60 As six-membered

chelate rings are more flexible than five-membered chelate rings, they can more

Introduction 19

readily satisfy coordination changes in the redox process. 61

When selecting ligands which will mimic histidyl-imidazole moieties, it is necessary

that the nitrogen donors be aromatic or, at least, Sp2 hybridised. As a result, imine

nitrogens are preferred to amine - nitrogens. Ligands containing imine nitrogens

can also stabilise copper complexes in the lower oxidation state better than amine

nitrogens, because the unsaturated ligands can delocalise electron density from

the metal ions via l7-back-bonding. 62 Heterocycles containing nitrogen are

particularly favoured.

Another aspect to be considered is the basicity of the donor group in the ligand,

since the reactions under investigation involve electron transfer. 58 A list of ligands

with their corresponding basicity is provid>~d in Table 6.

Table 6: N-donor groups and their p~ values.

Donor group p~

Benzimidazole 8.5

Imidazole 7.0

Histidine 8.0 J

Pyridine 8.7

Pyrazole 11.5

Alkyl-NH2 3.0

Aryl-NH2 9.4

The use of imidazole groups has been hampered by cyclisation, encountered in

the synthesis of ligands in which they have been included. Consequently,

derivatives such as benzimidazole have been used as donor groups as they are ~

less prone to cyclisation and, moreover, their increased steric bulk may be useful

in controlling the metal ion stereochemistry.63 A disadvantage, however, is that the

aromatic ring (fused fo-the imidazole ntlcleus) can affect the spectroscopic

properties of the resulting complex and mask the UV region, so important in

Introduction 20

studying copper(l) complexes. 58

Pyrazole groups have also been incorporated into ligands but, despite their easy

preparation64 and their spectroscopic similarity to imidazole, they are significantly

less basic. Pyridine groups have the advantage that their basicity is similar to that

of the histidyl-imidazole, and models in which pyridine has been incorporated have

been successful in mimicking the reactivity65 of copper proteins.

1.3.2 Coordination considerations in the reaction of copper complexes

with dioxygen.

During the binding of dioxygen to copper, two charge-transfer processes can *

occur, viz., metal-17-to-dioxygen-17 back donation and donation from oxygen to the

metal. This ultimately results in, at least partial, electron transfer to the metal and

weakening of the 0-0 bond. The binding of dioxygen to copper is also

accompanied by changes in both the oxidation state and coordination number of

the metal. The observed coordination numbers and stereochemistry is dependent

upon the oxidation state and, in the dO Cu(l) ion, the coordination number of four

dominates. The Cu(lI) cf ion, however, is the most prevalent oxi~ation state for

copper and, in this state, an octahedral coordination environment domil)8tes. In

this coordination environment, there.are four equatorial ligands that are strongly

bound and two axial ligands that are less strongly bound (Jahn-Teller effect). The

coordination numbers and geometries observed for the two oxidation states of

copper are summarised in Table 7.

The coordination numbers and geometries of the Cu(l) and Cu(ll) ions determine

the resulting redox reactions for the two oxidation states. The redox process which

occurs is also strongly affected by the type of ligand, the chelate ring size and the

- solvent. Six-membered chelate rings stabilise Cu(l),66-while five-membered rings

stabilise CU(II).67 When_polar solvents are used, the Cu(ll) oxidation state is

favoured but, conversely, non-polar solvents favour the Cu(l) oxidation state.68,69

Introduction 21

Table 7: Cu(l) and Cu(ll) coordination numbers and geometries70

ION Observed Observed geometry

coordination

number

Cu(l) 2 Linear

3 Trigonal

4 Tetrahedral

Cu(iI) 4 Square planar

Tetrahedral

5 Square pyramidal

6 Trigonal bipyramidal

Octahedral

Introduction

1.3.3 The mode of binding of dioxygen in oxyhaemocyanin and

oxytyrosinase

22

The dinuclear copper proteins, haemocyanin and tyrosinase, bind dioxygen

reversibly at their active sites. It is believed that these proteins have very similar

oxy active sites,42 involving two copper(lI) atoms (based on X-ray absorption

data71 ,16) and bound peroxide (based on the unusually low Raman 0-0 stretching

band72 ,73 at 750 cm-1). The studies showed that a peroxide ion bridges the

dinuclear copper active site of deoxyhaemocyanin and that oxidation of the metal

ions from copper(l) to copper(lI) occurs upon coordination of dioxygen. Intense

absorption bands due to CT transitions have been observed for both

oxyhaemocyanin and oxytyrosinase at 350 nm (8 = 20 000) and 570 nm

(8 = 1000). In a mononuclear copper-peroxige complex only two CT transitions are

observed.23 For oxyhaemocyanin, however, more than two peroxide-to-copper CT

transitions are observed, which is further evidence that the pair of copper atoms

are bridged. Because oxytyrosinase displays the same spectral features as

oxyhaemocyanin, its active site has been ascribed a similar structure to that of

oxyhaemocyanin.

, Research has been focussed on elucidating the structure of the peroxide ion

I

bridging the copper atoms. Initially, th~three end-on peroxide bridging structures

shown in Figure 5 were considered [(a) 1-1-1,1; (b) cis 1-1-1,2 and (c) trans 1-1-1,2J, but

only the cis 1-1-1,2 structure in (b) was expected to produce spectral features

consistent with those observed for oxyhaemocyanin.36 Several complexes have

been synthesised to mimic the cis 1-1-1,2 bridging in (b), but these were found to

have a trans 1-1-1,2 bridging mode and to exhibit spectral properties which differ

widely from those of oxyhaemocyanin. 74,75

"

Introduction 23

0 0-0 0-0 0

I / \ 0 / \ eu/ I "eu 0 eu eu eu/ \ eu Cu" /Cu c.f 'eu 0/

0 "0/ H

(a) 1-1-1,1 (b) cis 1-1-1,2 (c) trans 1-1-1,2 (d) cis 1-1-1,2 (e) l-I-r{1l2

Figure 5: Possible structural models for the bridging peroxide ion in dicopper complexes.

Strong antiferromagnetic coupling has been observed for oxyhaemocyanin and

oxytyrosinase, indicating that the copper ions are diamagnetic (EPR silent). To

account for this coupling, a cis 1-1 1,2- arrangement [Figure 5 (d)] involving an

endogenous bridge, originating from the protein, was proposed by Solomon. 76 The

two copper atoms were considered to coordinate to two or three histidyl-imidazole

ligands and were assumed to be in a copper(ll) oxidation state (Figure 6). The

endogenous bridge (OR) was thought to involve alkoxide, phenolate, sulphydryl

or hydroxyl groups in the protein or in the medium. The elucidation of the X-ray

crystal structure of haemocyanin77 from the spiny lobster (Panulirus interruptus)

confirmed both the dinuclear nature of the active site and the presence of three

histidine ligands per copper atom. However, the notion of an endogenous bridging

group was discarded as there was no suitable candidate within 12 A of t~ site. 78

N" .. I Io-q I .. "N CU Cu'

N/ I '0/ I "'N R

Figure 6: The dinuclear active site model proposed by Solomon. 76

A model that binds dioxygen in a similar way to that exhibited by oxyhaemocyanin

has been synthesised inthe laboratory of Nobumasa Kitajima. 79 The structure of

- the complex (Figure 7) showed that the dioxygen was bound in the novel 1-1-11 2 :112

bridging mode betweerr the two copper atoms [Figure 5 (e)].79 From the

spectrochemical and magnetic properties it could be seen that this model closely

Introduction 24

resembled the biosite in the protein. The controversy over the mode of dioxygen

bridging was eventually resolved when a crystal structure of oxyhaemocyanin from

the horseshoe crab was elucidated by Karen Magnus.8o A fJ-rt r)2 bridging mode

for dioxygen was, in fact, observed between the two copper atoms, each of which

is associated with three histidine nitrogens.

N, /01\ IN N-Cu Cu-N

I \ / '

NON

Figure 7: A model representing the fJ-r)2:r)2 bridging mode of a complex which emerged from the laboratory of Nobumasa Kitajima. 79

1.3.4 Synthetic biomimetic models of haemocyanin and tyrosinase

1.3.4.1 Early work

Before the X-ray crystal structure elucidation of haemocyanin~, research focussed

on developing systems exhibiting the spectroscopic or physical features which

characterise the biosite, the inference being that such systems reflect the structural

characteristics. Once a good synthetic model (based on spectroscopic and

physical features) was identified, then its X-ray crystal structure could be~sed to

elucidate the structure of the biosite.·

Thus, early research was based on an investigation of the physical and

spectroscopic properties of dioxygen-bridged systems. Key issues addressed in

this research included:- the structure of the deoxy-active site; the nature of the

protein bridge; the nature and geometry of the CU20 2 unit; the factors responsible

for the reversible binding of dioxygen; and the activation of molecular oxygen by

tyrosinase. 58

Introduction 25

1.3.4.2 Mononuclear Complexes

Before the discovery that the active site of haemocyanin was dinuclear, research

focussed on mononuclear copper complexes. The idea was that two mononuclear

complexes could be brought together by dioxygen to produce a structural model

of tyrosinase. Such systems were first reported by Wilson82-85 and later by

Casella,86 complex 1 being an example of the complexes developed by Casella.

R=H,C~H3

1

The red copper(l) complex 1 (R = H; L = pyridine) reacted with dioxygen in a 1:2

(Cu:02) stoichiometry at room temperature to give a green product, formulated as

a dinuclear copper(II)-peroxo species. This reaction may be partially reversed by

purging the solution with nitrogen.

- f ~ ~ I. _

-N N \\ 'I '\ .

~ /; 2 - 3

Complex 2, developed by Jacobson et a/. , 87 was one of the first completely

characterised complexes in which two mononuclear units are bridged by dioxygen.

The results of the spectroscopic analysis revealed that the structural features of

the complex, i.e. the Cu-Cu separation and the geometry around the copper ions,

were different to those in oxyhaemocyanin. It was also found that the peroxide

- bridge was trans rather than cis ~-1 ,2. The fact that the complex is diamagnetic

proves that an endogen9Y..s bridge is not a ~ecessary requirement for coupling of

the copper ions in the protein.

Introduction 26

The copper complex of the ligand 3 is dimeric in solid form, but mononuclear in a

solution of acetonitrile.88 The complex binds dioxygen in a 2:1 ratio (Cu:02) and,

at -BO°C, exhibits similar spectral features to those of complex 2. It has, therefore,

been proposed that the oxygenated dinuclear complex resulting from ligand 3 has

dioxgen bound in a trans 1-1-1,2 fashion. EPR measurements revealed that the

copper ions in the oxygenated complex are magnetically coupled (EPR silent).

4 ---- 5

The mononuclear complex 4 has been found to bind dioxygen reversibly forming

a I-I-peroxo copper(lI) complex 5;58 from the IR spectrum it is apparent that the

dioxygen is bound as peroxide. This reaction is reversible upon addition of

ethylene, but no structural data was reported.

Pyr 0 Pyr \ /1\ / Pyr--Cu Cu--Pyr / ,,/ \

Pyr 0 pyr

7

~N Pyr = I

The synthesis of the dinuclear copper complex 7, prepared from tris(pyrazolyl)

borate ligands 6 by Kitajima and co-workers,79 represented a major breakthrough

in this field. The electronic spectra, magnetic character, Cu-Cu separation as well

the I-I-peroxo IR stretching frequency were all very similar to those observed for

oxyhaemocyanin, and the 1-1-112: 112 bridging mode was confirmed by X-ray

crystallographic analysis:

Introduction 27

1.3.4.3 Dinuclear complexes

Systems based on the model proposed by Solomon 76 (Figure 6) were designed so

that a "reaction space" was provided in which the dioxygenated substrate could

be bound and activated. It was expected that the chemical reactivity and the

physicochemical characteristics of such complexes would be dependent upon the

coordination environment. The ligands were designed to coordinate two copper

atoms in sufficiently close positions to permit bridging by dioxygen.

R = H, Me

R' = H, CO:2Me

Complexes 8 and 9, which reflect this strategy, have been reported by Casella. 81

For complex 8 (R = Hand R' = H) oxidation of copper(l) to copper(ll) occurs when

the complex reacts with oxygen in dry acetonitrile. In methanol,however, the

oxidation of copper(l) is inhibited by the hydroxylation of the aromatic nucleus at , the C-2 position. In complex 9, hydroxylation at C-2 occurs on treatment with

J

oxygen, the resulting hydroxyl group acting as a bridge between the two copper

atoms.

~ HN~~ Ip' NH

ON-l-CU Cu- -N() I HN"~~ ~~ NH \

~b b~ 10 11

Complex 10 has been reported to react with dioxygen "semi-reversibly",81 while ~ ~ ~

complex 11, which differs only in the nature of the spacer, reacts with oxygen

Introduction

reversibly.

Hydroxide

14

Phenoxide and alkoxide

Alkoxide

(r ,~/

r-N. /0, /N, <.. Cu Cu ) N-N' 'I 'N-N V N3V

Phenoxide

28

13

15

In some cases, the ligand has been designed to have a hydroxyl group as part of

its structure, while in others an exogenous hydroxide bridge is inserted between

the copper atoms upon reaction with dioxygen. Several synthetic complexes (12-, 15) with different peroxide bridges are shown above. 58

Compared to their mononuclear analogues, dinuclear complexes have enhanced

reactivity and specificity in catalytic oxidation.89 Ligands that are binucleating have

the advantage of offering greater control over electronic and magnetic properties

than dimeric systems,90 and this has led to attention being focussed on the design

and synthesis of ligands that can coordinate two copper atoms. Karlin and co­

workers91 ,92 reported the synthesis of ligand 16; the resulting copper complex

reacts with dioxygen at !.ow temperature quaSi-reversibly. From EPR studies it

_ could be seen that the copper complex was dinuclear-and that it was EPR silent,

due to strong anti-ferrollJagnetic coupling be,!Ween the two cupric ions through the

peroxide bridge. It was suggested that the dioxygen was bound in a cis IJ-peroxo

or a bent IJ-112: 112 structure. 93 Although the crystal structure was not successfully

Introduction 29

determined, a Cu-Cu distance of ca. 3.4 A was estimated on the basis of EXAFS

data,93 and more recent studies have shown that the complex is virtually

diamagnetic. 94

Complex 17, reported by Zippel, displays antiferromagnetic coupling between the

two copper(II) atoms, bridged by alkoxy groupS.95 From the X-ray crystal structure

the Cu-Cu distance was found to be 3.033 A. Studies on this complex revealed that

the same structure is present in both the solid state and in solution, but that the

complex does not show any tendency to effect oxidation of c~techol.

Complex 18, reported by MOiler and co-workers,96 adopts an open conformation

resulting in a Cu-Cu distance of 6.16 A in the crystal state. No antiferromagnetic

coupling was observed for the two copper atoms because of the large Cu-Cu

separation. The formation of an oxygenated complex was, however, precluded by

the formation of a coordination polymer.

19

M ((N 0 N\'j py( \C~ \:~~-py

,,/\ ,'\/ py 0-0 py

Introduction

N: /":N 0 /N\l r( (\ / 'p~~) Py 20 Py~CU I o.i ,,/\! '7

Py 0-0'

30

Py = pyridine

The symmetric dinuclear complex 19, having a phenoxo bridging group, binds

dioxygen quasi-reversibly at low temperature yielding an intensely purple coloured

complex.97 The suggestion that dioxygen is a peroxo adduct is based on Raman

spectroscopy [v(O-O) = 803 cm-1] and dioxygen absorption stoichiometry (Cu/02

= 2) data. A Cu-Cu distance of 3.3 A for complex 19 was determined by EXAFS

analysis. Complex 20 is similar in structure to complex 19 but is unsymmetric; it

has been shown that this complex reacts ~ith dioxygen reversibly at -80°C and is

also intense purple in colour.97 The coordination of dioxygen to the copper ions

has been presumed to be the same in both complexes (19 and 20) because of

their spectral similarity. Complex 20 is, however, more stable than complex 19

and, as a result, it has been possible to isolate it in solid form.

21 22

The peroxo adduct @f complex 21, reported by Nishida et a/., shows

- antiferromagnetic coupling. 98 From the X-ray crystal structure, it is apparent that

the copper ions are Gogrdinated in a tri,gonal bipyramidal geometry with the

peroxide in an axial position. The copper(ll) ions in complexes 22 and 23 also

Introduction 31

exhibit antiferromagnetic coupling. 99,10o

R = H, Me, Et.

The dinuclear copper complex of the macrocycle 24 is a bright yellow solid that

changes to green when exposed to dioxygen in solution, and less rapidly in the

solid state. 15 It has been proposed that a ll-peroxo-dicopper(lI) species is

generated during the dioxygen uptake. Oxidative dehydrogenation occurs during

this dioxygen uptake. The structures an,9 properties of some dicopper(l) and

dicopper(ll) complexes of the macrocycle 25 have been reported. They function

as catalysts for the oxidation of several organic substrates, including catechols, in

the presence of oxygen. 101 In the [Cu2(25)(OEt}z(NCS}Z1 complex, each copper(lI)

ion is bonded to two imino nitrogen atoms of the macrocycle, to the nitrogen of

one terminally bound thiocyanate and to two bridging ethoxide groups in an

approximate trigonal bipyramidal geometry. The Cu-Cu distance is 3.003 A and the ~

furan oxygen atoms are not coordinated. The copper(lI) ions are strongly I

antiferromagnetically coupled in the bis(ll-alkoxo) complex and less strongly in the

dihydroxo complex. 15

26

_ Chiral dinuclear copper complexes have been relatively unexplored as catalysts

for enantioselective oxidations. Feringa has reported the chiral dinuclear copper ~ .-. ;>-

complex 26,102 which exhibits square planar geometry around each copper(lI) ion

and a normal Cu-Cu separation of 2.971 A. A trans-orientation of the N-benzyl-

Introduction 32

groups is found in complex 26 which, as a consequence, has C2-symmetry.

1.3.5 Monooxygenase Models

Since tyrosinase is a monooxygenase, a successful model of this enzyme would

be one that can catalyse the hydroxylation of an aromatic ring. Models reported in

the literature have done so in two ways, hydroxylating either an exogenous

substrate or the ligand itself. In the section that follows, attention will focus on

complexes that catalyse the hydroxylation of an exogenous substrate as well as

those that hydroxylate the ligand itself.

Oxidation of exogenous substrates

Several complexes have been reported th~t catalyse the oxidation of exogenous

substrates. Kitajima's fJ-peroxo complex 7 which resembles the tyrosinase enzyme

so closely in its spectroscopic features, has been reported to react with 3,5-di-t­

butyl phenol (DTBP) via a radical mechanism to afford couple.d products. 79

27 28

I

Reglier et a/. reported the oxidation of DTBP (2,4-di-t-butylphenol) in the presence

of triethylamine, by the biphenyl dinuclear copper complex 27.103 It is thought that

deprotonation of the phenolic substrate promotes binding to the catalyst. A coupled

product 28 is formed in the absence of triethylamine. Complex 29 (n = 4), which

reacts reversibly with dioxygen to form diamagnetic peroxide bridged structures, 92

has also been reported to give the coupled product 28 via a radical mechanism.

Introduction 33

29

Complex 30 has been reported to catalyse the oxidation of DTBP to the quinone,

and the phenolic substrates 31 to the dihydroxyphenols.104 Evidence for the initial

formation of an intermediate phenolate adduct was obtained.

30

A !VIe r----:N N~!VIe

6'N\\ I "Cu cl '\ ~ ":oN ~N \ / \ / N# I ):::::-N N~ \

A Me--N6 oN-Me "'--

(r0H R 31

R = p-CO:zCHa or DTBP

Complexes of the macrocyclic ligand 32 catalyse the oxidation of exogenous

substrates, 2,6-di-t-butylphenol being oxidised to the coupled product 33. , I

32 =0=0= o 0 - -

33

Reim et al. have reported symmetrical and unsymmetrical dinuclear copper

complexes that catalyse the oxidation of 3,5-di-t-butylcatechol to the o-quinone. 105

The complexes were pr.epared from ligands 34, 35, 36 and 37. The dinuclear

_ copper complex prepared from the symmetric ligand 34 displayed the highest

catecholase activity of pll the dinuclear copper complexes prepared from these

ligands.

--

Introduction 34

Br Br

A A 34 eN) OHeN) eN) OH HN 35

N N ~d I I Me Me Me _ Br Br

A ~ 36 eN) OH HN eN) OH HN 37

l d l J ~N-Me

Hydroxylation of the aromatic ring of the ligand.

From studies conducted on the hydroxylation of the aromatic ring in dinuclear

copper(l) complexes, it seems that the reaction is ligand dependent. Sorrel and his

co-workers prepared complexes 38 and 39 and investigated their reaction with , dioxygen. 58 They found that these copper(l) complexes neither catalyse the

J

hydroxylation of the aromatic ring nor react with dioxygen in dichloromethane.

A A R

~ ~ 38 H Pyr = I ;;

/N 39 CH3 (01u TCJJ (0eu eCJ;

40 R

Py = pyridine Pyr' I I ./ Pyr Pyr'Pyr Pyr Pyr Py Py

Introduction 35

However, when the mixed pyrazole-pyridine copper(l) complex 40 was treated with

dioxygen, a reaction was observed, but no aromatic hydroxylation occurred.

41 42

Dehalogenation reactions are catalysed by various enzymes using hydrolytic,

reductive or oxidative pathways,106 but only a few enzymes are capable of

oxidative dehalogenation of aromatic compounds. The dinuclear copper(ll)

complex 41 has been reported to undergo oxidative debromination to afford

complex 42 upon reaction with dioxygen at room temperature.

A M \ 1

eN>" rf,~ eN 0 J 43 ' I \ / 44 Cu Cu / \ I "'-

R 'R R 0 R

J

Casella and Gullotti have reported mOrJooxygenase activity for complexes of type

43, which are oxidised to the corresponding complexes 44. 107 The hydroxylation

of the ligand is affected by the solvent, the oxygenation reaction being partially or

completely depressed in dry acetonitrile. Protons have been found to enhance the

hydroxylation of some of these systems and, as a result, it has been suggested

that an electrophillic copper-peroxo or copper-hydroperoxo complex is the active

species in the hydroxylation process. A proposed mechanism for the hydroxylation

reaction is illustrated in-5cheme 1, and low-temperature spectral studies have

- been undertaken to characterise the intermediates. 107 -

Introduction 36

M m m 02 H I w I ~ I N H N ---.. ---..

(' I II ) N - N N H, N Cu Cu / \ (' "t III) C ' "f "I ) N N CU- -CU CU- -CU

/ - \ / - \ NON NON

1

MI MI A NON ...-- NON ...-- NON C ' 11/ , IV ) ( \ III , DI ) C ' "I ' "I ) Cu Cu Cu Cu Cu Cu

/ \ / \ / \ \ / \ \ NON N OH2 N NON

; \ , \

H H H H

Scheme 1

Casella and co-workers have also reported aromatic hydroxylation to give the

dinuclear copper(l) complex 45, containing methionine sulfur groupS.108 The .

dinuclear copper(lI) complex 46 forms as a result of hydrolytic clea~age of 45, and

has been characterised by X-ray crystallography. J

o /(l?11

o NON 0

45 I '/' / I

o--C�-o------Cu ~------O--Cl-O 46 I / \ / \ I o NON 0

I~I 17, ~

Introduction 37

1.4 THE IMPORTANCE OF METAL-DIOXYGEN BONDING

The flourishing interest in dinuclear and polynuclear complexes emanates from the

fact that pairs or clusters of metal ions have the capacity to mediate certain

chemical reactions of industrial significance either more efficiently than, or in a

different manner to, isolated metal centres. Dinuclear complexes having metal

centres in close proximity, have been the theme of much recent research since

these structural units are believed to be involved in an array of crucial biochemical

processes, especially oxygen transport and oxygen activation by metal - containing

proteins and enzymes. "l()9 .. 120 Reactions in which one or both atoms of dioxygen are

catalytically inserted into an organic substrate are of significance in the synthesis

of metabolic products and intermediates. Dioxygen also serves as an electron sink

in the oxidation of a variety of small molecul~s such as ascorbic acid, catechol and

amino acids. 121,122 Complexes, both synthetiC and natural, which bind dioxygen

reversibly have been termed oxygen carriers. In addition to their significance as

models for natural oxygen carriers, synthetic dioxygen complexes have potential

applications in dioxygen separation and storage, industrial processes and

catalysis. 122 These complexes may act in catalytic or stoichiometric fashion. The

concepts originating from these studies are expected to contribute to the evolution ~

of practical synthetic systems for the reversible binding of oxygen and~or the

oxidation of organic compounds.

The reduction of dioxygen is of major importance for biological systems and for

oxygenation reactions. The free energy change for the four-electron reduction of

dioxygen to two water molecules is -316 kJmol-1 at pH 7 (eq. 1).

O2 + 4H+ + 4e- --. 2H20 (eq. 1)

P = +0.81 §.V (vs. the normal hydrogen electrode)

The reduction does no( os;cur in a single ~tep, but rather rn a series of steps

involving successive single - electron transfers. The usefulness of the reduction

Introduction 38

is, however, severely restricted. A common pathway for the reduction of dioxygen

involves a one-electron reduction followed by disproportionation to form OH­

(under basic conditions) or H20 2 (under acidic conditions). The effective potential

is pH independent and the single-electron reduction is endothermic by 128

kJmol-1.

It has been found that the oxidation of organic compounds is more efficient when

dioxygen is directly involved with dinuclear rather than mononuclear complexes.

The formation of metal hydroperoxides are of fundamental importance during these

oxidations and, as a result, the formation of such species has been investigated.

Copper is a metal which is particularly important in oxidation catalysis. It has been

used extensively in the laboratory and in in~ustry. The synthesis of acetaldehyde

by catalytic oxidation of ethylene, the so-called Wacker process, is of major

importance in the manufacture of organic chemicals from ethylene. The reaction

is shown in equation 2.

(eq.2)

~

Although palladium is the main catalytic component responsible for the oxidation J

of acetaldehyde in the Wacker process, copper is the most effective co-catalyst to

re-oxidise Pdo to P&+with dioxygen.123 The oxidative coupling of acetylene to give

diacetylenes, the Glaser reaction, is another well-known reaction catalysed by

copper (eq. 3). This is synthetically useful124 because many ethynyl compounds

are coupled almost quantitatively. The coupling is effected by bubbling oxygen

through an acetone solution of acetylene in the presence of a copper(l) chloride

tetramethylethylenediamine complex. 125 This reaction is important in the

commercial dimerisation~of acetylene, which is then used as a precursor for

_ chloroprene, the monomer for neoprene rubber. 60

Introduction 39

~ 2 R-G=C-R - R-G=C-G=C-R (eq. 3 )

Aerobic oxidation of 2,6-xylenol with copper(l) in the presence of a N-donor ligand

produces a p-phenylene oxide polymer which is widely used as an engineering

thermoplastic under the trade name PPO (eq. 4).126

(eq. 4)

n

Ethylene glycol is produced directly from ethylene by oxidation in the Teijin

process which utilises a CuBr system as. a catalyst (eq. 5).127,128 This oxidation

process is important since ethylene glycol is a major intermediate in the synthesis

of polyesters and polyurethanes.

(eq. 5)

The copper-mediated oxidative decarboxylation of benzoic acid to\ form phenol is

an industrially useful reaction (eq. 6) as is the catalytic cleavage of catecjlOl (eq.

7).129,130

OH

1/2~ • 6 +~ (eq.6)

(XQ-i

. I ~ Q-i

(eq. 7)

Introduction 40

Acrolein is manufactured by oxidising propene in the presence of heterogeneous

catalysts containing copper or cuprous oxide, while methanol is produced from

synthesis gas by a non-oxidative heterogeneous process using a CuOIZnO

catalyst. Hydroquinone, an important commodity chemical primarily used as an

antioxidant and in photographic chemistry, is produced by the copper-catalysed

oxidation of phenol with dioxygen to afford p-benzoquinone, which is then

reduced. 131 The commercial production of trimethylhydroquinone, a vitamin E

precursor, is based on the copper-catalysed oxidation of 2,3,5-trimethylphenol to

the benzoquinone. 131 It is believed that copper-dioxygen complexes play important

roles in these catalytic oxidation reactions. However, relatively few intermediates

have been isolated and characterised. Furthermore, the structural factors which

determine the catalytic effectiveness of the catalysts are not clearly understood.

There has also been considerable interest in the reactions of dioxygen and

hydrogen peroxide with copper(l) complexes of 1,1 O-phenanthroline (phen),

[Cu(phen)2t and substituted phen derivatives, following .the discovery that

[Cu(phenht cleaves deoxyribonucleic acid (DNA) in the presence of hydrogen

peroxide.so

I

Introduction 41

1.5 OBJECTIVES OF THE PRESENT INVESTIGATION

The primary aim of this study has been to design and prepare synthetic models of

the enzyme, tyrosinase. In previous work in our group, synthetic models which

contained an imine functionality were synthesised. 138 These synthetic models

displayed some biomimetic activity but, unfortunately, the pure complexes could

not be isolated, thus precluding unambiguous identification. It was suspected that

the imine functionality was involved in intramolecular cyclisation resulting in

formation of mixtures. In order to circumvent the problem of intramolecular

cyclisation, attention in this study has been focussed on ligands which do not

contain the imine functionality, and the following objectives were identified.

1. The design and synthesis of novel ligan9s capable of forming dinuclear copper

complexes and acting as biomimetic models of the active site in tyrosinase.

2. The preparation and characterisation of copper complexes of these ligands.

3. An investigation of the copper oxidation state(s) in the re.sulting complexes

using cyclic voltammetry.

4. Evaluation of the phenolase and catecholase activity of the copper

complexes using 3,S-di-t-butylphenol and 3,S-di-t-butylcatechol as \

substrates respectively. , 5. An investigation of complexes of the ligands with other metals, such as

cobalt, nickel and platinum.

--

42

2 DISCUSSION

In the discussion that follows attention will be focussed on: - the synthesis of

biphenyl ligands (Section 2.2.1); 1,1 O-phenanthroline ligands (Section 2.2.2);

Schiff base ligands (Section 2.2.3); ligands prepared via the Baylis-Hillman

reaction (Section 2.2.4); macrocyclic ligands (Section 2.2.5); and a dendrimer­

based ligand (2.2.6). This is followed by complexation and computer modelling

studies (Section 2.3). Electrochemical data, which help to elucidate the oxidation

state(s) of the metal in copper, cobalt and nickel complexes, are considered in

Section 2.4. Finally, the biomimetic evaluation of selected copper complexes is

discussed in Section 2.5.

2.1 LIGAND DESIGN AND RATIONALE-

The nature of the hydrocarbon groups (the "spacers") linking the coordinating units

i.n dinuclear copper complexes had to be taken into account as this can markedly

affect the reactivity of the copper complexes. 132 The ligands were designed,

primarily, to incorporate copper(l) and copper(ll) ions (Figure 8).

1\ 1\ N N-R N, , /N-R

'Cu'

,Cu, N N-R N/ 'N-R

~ ~

(a) (b)

Figure 8: (a) The generalised ligand structure; (b) the complex after complexation with copper, but prior to oxygen bridging.

.1 R

0 ~I

N~ LNH

NJ:NH

0

Discussion 43

In these ligand systems, the spacer was varied to explore the effect of complex

structure and geometry on biomimetic activity. For each of the ligands examined,

there are at least two different types of nitrogen donor, one from an amide, amine

or imine functionality attached to the "spacer", and the other from a heterocyclic

system (R = pyridine, imidazole or benzimidazole; Figure 8). The electronic

characteristics of the different nitrogen donors provide for variation in electron

availability at the donor sites in the different complexes.

The incorporation of different nitrogen donor types is justified106 by the fact that, in

haemocyanin, one histidyl imidazole is situated further away from the copper ion

than the other two; 133 it is believed that this is also likely to be the case in

tyrosinase. 134 Additional support is provided by the activity shown by complexes

containing different donor types.13S.136 TheJigands have, typically, been designed

to form 6-membered chelate rings on complexation, because this arrangement is

stable for copper complexes and permits changes in the oxidation state. 61 For the

ligands containing the planar 1,1 O-phenanthroline spacer, however, formation of

a 5-membered chelate ring during complexation with copper is possible - an

arrangement known to favour the stabilisation of the copper(lI) oxidation state. 137

The substrate is expected to gain entry to the copper active site, i~ each case, by

moving between the arms of the ligand. I

In our attempts to develop synthetic models of the active site of the enzyme,

tyrosinase, it was decided to prepare and isolate the organic ligand prior to

complexation rather than use the template approach. This allows for the

independent isolation and characterisation of the ligand without any interference

from the metal ion.

Discussion 44

2.2 LIGAND SYNTHESES

2.2.1 Biphenyl ligands

The biphenyl unit has been used as a spacer to impart flexibility to the organic

ligand and the resulting complex.138 This spacer allows the complex to adopt a

conformation in which the two copper atoms are sufficiently close together to

permit bridging by dioxygen without imposing rigid planar geometry. It has been

shown that if the spacer is too flexible, binding of the substrate may be inhibited. 139

The synthetic routes used to prepare a series of ligands containing the biphenyl

spacer are outlined in Scheme 2. Biphenyl-2,2'-dicarbaldehyde 48140 and biphenyl-

2,2'-dicarboxylic acid 49141 were produced from phenanthrene 47 in separate

reactions following reported methods. Thus, ozonolysis of phenanthrene 47 in

methanol at low temperature [to inhibit formation of a stable peroxide 50 from the

reaction of the intermediate with methanol (Scheme 3)] afforded biphenyl-2,2'­

dicarbaldehyde 48 in high yield (93%). The 1H NMR spectrum (Figure 9) of the

dicarbaldehyde 48 is charactersised by the aldehydic proton signal at 09.83 ppm.

J J

I' , 'I' PPIII 10 9

48 g I ~I CHO

OHC I~

b

Figure 9: The 400 MHz 1H NMR spectrum of biphenyl-2,2'-dicarbaldehyde 48 in CDCI

3. ~ -

Discussion 45

Access to the dicarboxylic acid 49 was achieved via H20 2 oxidation of

phenanthrene 47 in glacial acetic acid, the required product being isolated in 67%

yield; the presence of the carboxylic acid group was confirmed by a 1 H NMR signal

at 012.44 ppm and a broad IR absorption band in the region 2500-3500 cm-1.

< > < > 0 3

OHC RNH2 KI Q (> 47 - 48

MeOH NH2

CHO 51

~02 t C~C02H

t RNH2 CHCI3 6 ::::,...1

49

55a-c

NR

H02C NH2

< ~ ( > 54a-c H~N 52 C02H 'I RN

t RNH2 t NaBH4 COl MeOH DMF

NHR NHR 53 UAIH4

* 56a-c

THF RHN RHN I

a b c

Scheme 2: Synthetic routes for obtaining ligands containing the biphenyl spacer.

0

~ \LNH

Discussion 46

OHC KI OHC

MeOH CHO

47 MeOH~ 48

CH-9 50 CH-O

\

OCH3

Scheme 3: Synthesis of biphenyl-2,2'-dicarbaldehyde 48.

The chelating ligand system, 2-(2-aminoethyl)pyridine 51, was available com­

mercialy, while histamine 53 had to be released from its dihydrochloride salt by

treatment with sodium methoxide. 2-(2-Aminoethyl)benzimidazole 52 was obtained

by reacting 1 ,2-diaminobenzene 57 with p-alanine 58 (Scheme 4).138 The formation

of the bis-substituted biphenyl derivatives 54a, 54b and 54c containing these

chelating systems, however, presented particular difficulties.

~N~ + ~N~

57 58

6MHCI .. 52

Scheme 4: Preparation of 2-(2-aminoethyl)benzimidazole 52.

J

The formation and subsequent reduction of the diimino derivatives 54a, 54b and

54c of the dialdehyde 48, was expected to afford the required di-amin9 ligands

56a, 56b and 56c. The tendency of imidazolyl imines to undergo intramolecular

cyclisation, observed in our group 138 and reported by Casella ef a/. 135 (Scheme 5),

had prompted us to target the diamines 56a, 56b and 56c as model ligands rather

Discussion 47

than their di-imino precursors 54a, 54b and 54c. However, the reduction step

proved more difficult than expected.

Scheme 5: An intramolecular cyclisation reaction proposed by Casella et al. 135

The di-imino compound 54a, containing pyridyl groups, was prepared using the

procedure employed by Reglier et al. 103 Attempted reduction of this compound

using sodium borohydride in methanol at room temperature, however, failed to

afford the expected product, 56a. The 1H NMR spectrum of the isolated product

(Figure 10) are clearly inconsistent with structure 56a. It was expected that the

signals for the methylene protons at 02.98, 03.13 and 03.48 ppm would each

integrate for four protons but, instead, the integrals at 02.98 and 03.13 ppm are

equivalent to two protons each, while only the signal at 03.48 ppm integrates for

four protons. The 13C NMR spectrum (Figure 11), however, reveals the expected

14 signals (2 overlapping at ca. 55.5 ppm; 3 at ca. 127.5 ppm). \

f"

ppm '0 . ~ 'I' 'I • i. I

7 ,

61a

'I' 5 I""" to 4 3

fr

'I 2

I"" 'i ., , 0

Figure 10: The 400 MHz 1H NMR spectrum of ligand 61a in CDCI3.

I

.'" " " "~a 1Jto I ,60

I

I 140

I

Discussion

i

I

'" I

'00

......

~'

Figure 11: The 13C NMR spectrum of ligand 61a in CDCI3.

I .0

• "1' o

48

When the reaction was repeated at room temperature, the same product was

isolated. Single crystal X-ray analysis (see sections 2.3.4.1.2, p.1 07, Figure 36 and

2.3.4.2.2, p.122, Figure 46) of the cobalt and nickel complexes of this compound

revealed that the ligand was, in fact, the dibenzoazepine (R = pyridine; Scheme ~

54), thus explaining the apparent anomalies in the NMR spectra. I

Formation of ligand 61a may be rationalised in terms of the mechanism outlined

in Scheme 6. It is suggested that one imine functionality is first reduced during the

reaction with NaBH4 . The lone pair on the nitrogen of the resulting amine

functionality 59a then attacks the remaining imine functionality, resulting in

cyclisation and subsequent loss of one arm of the ligand from the intermediate

aminal 60a. Similar reduction of diimines 54b and 54c also failed to yield the

expected products 56b and 56c, affording, instead, the corresponding

dibenzoazepines 61b and 61c. One- and two-dimensional analysis (illustrated for

compound 61 b in Figu~es 12 and 13) pr?vided unambiguous confirmation of the

identity of these cyclised products.

Discussion 49

R ~R NaBl-i4 ..

MeOH a () ~I

54a-c 59a-c HNtN

1 b U C ~N

61a-c N~R HN..!I •

Scheme 6: Proposed mechanism for the intramolecular cyclisation during NaBH4 reduction of the diimine 54a.

~I

61 b ~ N-~-.tD I N~

b

I

\ ----)'----~

pp.

Figure 12: The 400 MHz 1H NMR spectrum of ligand 61b in MeOH-d4.

, -

o

Discussion

, <. , < .•

01

(l 10

Figure 13: The HETCOR spectrum of ligand 61 b in MeOH-d4.

50

In an alternative approach, it was decided to focus attention on ligands containing

amide rather than imine functions and to effect reduction to the corresponding

amine ligands 56a, 56b and 56e using lithium aluminium hydride (Scheme 2). The \

diamides were prepared by reacting the diacid 49 with the coupling agent carbonyl I

diimidazole (CDI),142 at 4(1)C and then treating the resulting intermediate with the

amines 2-(2-aminoethyl)pyridine 51, 2-(2-aminoethyl)benzimidazole 52 and

histamine 53. The diamides were obtained in 54-82% yield and were fully

characterised by elemental (high resolution MS) and spectroscopic analysis.

Broadening and coalescence of certain signals is apparent in the 1 H NMR spectra

of ligands 55b and 55e. These effects are illustrated in the 1H NMR spectrum of

diamide ligand 55e where broad peaks are observed at 02.51 and 03.34 ppm

(Figure 14). The broadening of the peaks are attributed to slow rotation of the

- bonds on the NMR time scale. Unfortunately, numerous attempts to reduce these

diamides (55a, 55b and 55e) with reducjng agents such as lithium aluminium

hydride and Raney Nickel were unsuccessful.

Discussion 51

55c WI H

FN O~ ~ HN~ 0 ~NH

~ I ~ A

- J

Figure 14: The 400 MHz 1H NMR spectrum of diamide ligand 55e in MeOH-d4.

In a third approach, a synthetic route reported by Sorrell and Garrity,143 was

explored. These authors had also been unsuccessful in their attempts to reduce

similar amide derivatives and, as a result, they developed a synthetic route that

involved initial protection of the amine as an amide, followed by de\protection using

acid hydrolysis. A similar synthetic approach (Scheme 7) was investigated for

generating the diamine ligand 56b. "The amide 62 was obtained by condensing

2-(2-aminoethyl)benzimidazole 52 with benzoic acid and using COl as the coupling

agent. This amide was then treated with NaH and 2,2'-dibromomethylbiphenyl to

give the diamide ligand 63. The structure of the ligand was confirmed by elemental

and spectroscopic analysis; the complexity of the aromatic region in the 13C NMR

spectrum can be clearly seen in Figure 15. Attempts to remove the benzoyl

protecting groups by ac:.id hydrolysis proved unsuccessful. Given the unexpected

difficulties in accessing the diamino ligands, we decided to examine the various

amide ligands, which had been obtained, as potential ligands forcomplexing

copper (see section 2.3.2).

Discussion 52

CO:zH

>-0 O:N~NH' 0 ~ O:yN -~ NH DMF I" H

COl ~ NH 62 52

Br

Nr;; qb DMF N2

Br

~ -- 0

<~~ ~i> 63

o~ HN ~ h

~ 6N Hel reftux 10N NaOH

H

Ji> (~~N J 56b

N HN H

Scheme 7: Attempted synthesis of ligand 56b.

Discussion 53

Figure 15: The 400 MHz 13C NMR spectrum of ligand 63 in o fv\SO-d6·

Discussion 54

2.2.2 1,1 O-Phenanthroline ligands

Ligands containing the 1,1 O-phenanthroline unit as spacer were expected to exhibit

some rigidity and planarity. Such rigidity in copper complexes may well favour

binding of the substrate and, thus, enhance their biomimetic potential. The

synthetic routes followed to obtain 1,1 O-phenanthroline-containing ligands are

outlined in Scheme 8.

Neocuproine 64 (2, 9-dimethyl-1 ,1 O-phenanthroline) was used as the starting

material. It has long been known that this compound is a useful analytical reagent

for the determination of copper, 144 forming a bright red copper complex. The Se02

oxidation reactions of neocuproine 64 to 1,1 0-phenanthroline-2,9-dicarbaldehyde

65145 and the subsequent oxidation to 1, t.o-phenanthroline-2, 9-dicarboxylic acid

66 145 using 80% HN03, occurred with relative ease. The 1 H NMR spectrum of 1,10-

phenanthroline-2,9-dicarbaldehyde 65 (Figure 16) reveals the characteristic

aldehydic proton signal at 010.35 ppm.

Condensation of 1,1 0-phenanthroline-2, 9-dicarbaldehyde 65 with the amines 51

and 52 gave the corresponding imines 67a and 67b which, without further

purification, were reduced using NaBH4 to afford the required diamino ligands 68a

and 68b. These ligands were fully characterised, and 1- and 2-D NMR spectra for

compound 68a are illustrated in Figures 17 and 18 respectively. From the 1H NMR

spectrum it can be seen that the formation of the diamine ligand 68a was

successful, the singlet at 04.32 ppm corresponding to the methylene groups. The

HETCOR spectrum of ligand 68a (Figure 18) reveals the requisite correlations

between the carbons and their attached protons.

Discussion 55

0=0 0=0 RNH2

Se02 64 .. 65

1,4-dioxan

6 H3C CH3 OHC CHO 51

% 1 RNH2 80%HN0.3 :::::,...'

~ 0=0 NH2

67a,b 66 H~N 52 HOzC COzH

NHR RH\J 0 t NaBH4• 1 RNH2• MeOH COl,

DMF

~ ~ LiAIH4 53

68a-c -->< 69a-c THF

0 \LNH

0 NHR RH\J NHR RHN

abc

I

Scheme 8: Synthesis of ligands containing the 1,1 O-phenanthroline spacer.

Discussion

0=0 OHC CHO

65 j[-

j Iii i I I I I i j I ! I I I I I iii j i i [ I I , j I , , I f I I Ii! I i I Iii f

ppm 10 ~ B 7

Figure 16: The 400 MHz 1H NMR spectrum of 1, 10-phenanthroline-2,9-dicarbaldehyde 65 in DMSO-d6.

I' ... • I~

68a

~ ,

HN J)

\

" I NHU ~

1 ~ , I " 1 ' .~~

I 0

Figure 17: The 400 MHz 1H NMR spectrum of ligand 6Sa.in CDCI3 .

J

56

Discussion 57

J l 1 , ,

I

. I

~ ,I ,

~ I II

- I ~""'-

.- , , , , , , ,

Figure 18: The HETCOR spectrum of ligand 68a in COCI3 .

In an alternative approach, the dicarboxylic acid 66 was conden~ed with each of

the amines, 2-(2-aminoethyl)pyridine 51, 2-(2-aminoethyl)benzimidazo~ 52 and

histamine 53 using COl as a condensing agent. The resulting diamides 69a, 69b

and 6ge were isolated in yields ranging from 61 to 73% and, in each case, a signal

corresponding to the amide proton was observed at 09.5-9.8 ppm in the 1H NMR

spectrum (illustrated for ligand 69b in Figure 19). Access to the 1,1 O-phenan­

throline-based diamino ligands 68b and 68e was then explored via the reduction

of the intermediate diamides 69b and 6ge. Amide carbonyl groups are commonly

reduced using excess LiAIH4. Unfortunately, as was the case with the biphenyl

analogues, attempts to reduce the diamides 69b and 6ge using LiAIH4 were -

unsuccessful. It was decided to evaluate the diamides as potential ligands for

copper, the diamino analogues being available via the diimines.

69b

'" I DP'" 10

I' 7

Discussion

100.0 i '&

6 5 "~ o ~ ,.

Figure 19: The 400 MHz 1H NMR spectrum of ligand 69b in DMSO-d6 o

58

I

Discussion 59

2.2.3 Schiff base ligands prepared from diketones

A series of novel diimine ligands were targeted, which contain acyclic spacers and

which illustrate the effect of varying the length of the spacer chain and the

substituents R\ R2 and R' (Figure 20). As imidazole-containing imino ligands are

prone to intramolecular cyclisation, 2-(2-aminoethyl)pyridine 51 was chosen to

form the coordinating arms of the ligands. Various diketones were used to provide

the required spacer moieties. Furthermore, the ligands were designed to complex

copper via 6-membered chelate rings (see Figure 20b), thus allowing changes in

the oxidation state of copper.61 Additional advantages are that both the imine and

pyridine moieties are able to stabilise copper in the lower oxidation state.

R2 "" ... R2

-A~ RAR' R1 R2 R3

76 Me H Me 77 Me Me Me

cf ~ 8 9 78 Me Et Me Cu Cu

~ 79 Me Ph Me / "-_N N " 80 Ph H Ph

~ N N~ ~ B '1_ 81 Ph H Me (a) (b)

\ Figure 20: (a) Proposed Schiff base ligands and (b) their envisaged complexation

with copper. I

The synthesis of these ligands is outlined in Scheme 9. With the exception of the

diketone 73, which was prepared by the standard benzylation of acetyl acetone

(Scheme 10), the required diketones were available commercially. Reaction of

these diketones with 2-(2-aminoethyl)pyridine 51 gave the expected diimines 76-81

in yields ranging from 25 to 61 %. Similar condensation of the diketone 82 afforded

the diimine 84; numerous attempts to prepare the diphenyl analogue 85 from the

diketone 83 proved ~unsuccessful. The ligands were purified by flash

chromatography and characterised by NMR spectroscopy and by FAB-MS

analysis.

82

R2

RylyR3

0 0

70 -75

H2NU ~ I ~

51

70 71 72 73 74 75

0

H3C~CH3 0

CHCb

Discussion 60

R2

CHCb RhR3

dN

N~ ~ N N~

R1 R2 R3

Me H Me Me Me Me Me Et Me Me PHCH2 Me Ph H Ph Ph H 'Me

51

H2NU ~ I ~

76 - 81

76 77 78 79 80 81

0 0

Ph~Ph 83

I

CHCb

Ph~Ph

""i jN N'( "'" C~ fJ

85

Scheme9: Synthesis of Schiff base ligands.

Discussion 61

0 0 NaOMe 0 0

AA ~ Y 70 P~Br 73

Scheme 10: Synthesis of the diketone 73.

The 1 H NMR spectra for ligands 76-81 in which the imine functionalities are

separated by one carbon atom, reveal extensive tautomerism. A broad signal in the

region 012-13 ppm characterises the 1H NMR spectra of these ligands (illustrated

for ligand 78 in Figure 22) and is attributed to the hydrogen-bonded amino proton

in the enamine tautomer (Figure 21). This interpretation is supported by deuterium

exchange data. After the addition of 0 20 to a sample of ligand 78 in COCI3, the 1H

NMR signal at 012.37 ppm virtually disa'ppeared (Figure 23); furthermore, the

methylene signal at 03.65 ppm (observed as a quartet before deuterium exchange)

appeared as a triplet indicating coupling between the amino proton and the

adjacent methylene protons. In the 1H NMR spectrum of ligand 84 (in which the

imine groups are separated by two methylene groups), however, the signal at ca.

012 ppm is absent and the methylene signal at 04.13 appears as a triplet, not as

a quartet (Figure 24). Thus, it may be concluded that significant taMomerism only

occurs when the imine functionalities are separated by one carbon atom~ This is

not surprising since, only in these systems, is the enamine tautomer (Figure 21 )

stabilised by extensive conjugation and by a six-membered, H-bonded chelate.

Figure 21: Tautomerism in the Schiff base ligands 76- 81.

Discussion

r rI '" I"'" """ I'

pp" t2 to

J'J "j

4 1'0' Oi'I' 2 o

Figure 22: The 400 MHz 1H NMR spectrum of ligand 78 in COCI3 _

I 00.

12 1

10

r r r J Jj j J J

.. J.1J_ ... J. 11-,.~,------, r

0" 'I 4 , & " 'A '~ ,

Figure 23: The 400 lVI~z 1H NMR spectrum of ligand 78 in COCI3 after the addition of 0 20 _ <

62

Discussion 63

~CyyCH3

d, N \"---1 ~N 84 fJ

iii. i pp. !O

Figure 24: The 400 MHz 1H NMR spectrum of ligand 84 in CDCI3 _

The Schiff base ligands, derived as they are from acyclic diketone spacers, are

flexible, since rotation about the carbon-carbon bonds separating the two imine \

functionalities is possible. As a result, their copper complexes may be ex~ected to

readily adopt appropriate conformations for binding a substrate molecule_

However, it was hoped that the copper complexes would not be too flexible as this

could actually inhibit the binding of the substrate and result in an absence of

catalytic activity_

Discussion 64

2.2.4 The Baylis-Hillman approach to ligand synthesis

A further approach to the synthesis of ligand systems involved use of the Baylis­

Hillman reaction, 146 in which activated alkenes (typically, acrylate esters) react with

aldehydes in the presence of a nucleophilic catalyst, such as DABCO

{diazabicyclo[2.2.2]octane}, to afford multifunctional products 86 (Scheme 11). The

reaction has been the subject of several reviews,147 and has been applied in our

research group to the synthesis of heterocyclic systems, viz., indolizine,148

quinoline, 149 coumarin and chromene derivatives. 150 In the present study, it was

anticipated that application of the Baylis-Hillman reaction to a dialdehyde (e.g.

1, 10-phenanthroline-2,9-dicarbaldehyde 65; Scheme 12) would provide access to

polydentate ligands such as 89 and 90. Thus, reaction of the acetylated Baylis­

Hillman product 88 with piperidine or pyrrolidine was expected to lead, via a

conjugate addition-elimination sequence, to the ligands 89 and 90 respectively.

Before applying the Baylis-Hillman reaction to the dialdehyde 66 it was decided to

use pyridine-2-carboxaldehyde 91 (Scheme 13) as a model system to explore the

feasibility of the proposed transformations.

+ '\.-R' DABCO

86

Scheme 11: The Baylis-Hillman reaction.

R' = CN, OO;zMe

Discussion 65

~~ <~ OHC CHO

DABCO

87 * ~CN HO OH

65 NC CN

.------ 88 AcO OAc

NC CN

o 89

90

o o I

Scheme 12: Proposed Baylis-Hillman pathway to polydentate ligands.

2.2.4.1 Pyridine-2-carboxaldehyde as a substrate

The Baylis-Hillman product 92 was obtained after 4 days by reacting pyridine-2-

carboxaldehyde 91 with acrylonitrile in the presence of DABCO (Scheme 13).148

- Once this product had been obtained, two routes to compounds 96 and 97 were

considered. In the first, rE?aftion of pyrrolidine< or piperidine with the Baylis-Hillman

product 92 was expected to give derivatives 94 and 95 via an addition reaction,

Discussion 66

subsequent dehydration affording 96 and 97 respectively. Alternatively, acetylation

of the Baylis-Hillman product 92, followed by allylic (SN') displacement of acetate

from the acetylated intermediate 93 by piperidine or pyrrolidine, was expected to

afford compounds 96 and 97 respectively. In the event, the latter synthetic route

involving the acetylated product 93 was attempted first and was found to work well;

as a result, the alternative route was not attempted.

0 ~CN -P AC20 -P .. ... 93 DABCO HO O.5h reflux AcO

OHC CHCI3 4d NC NC

91 92 ,

RH I

THF 1 RH I ,

THF : stir for 3d [-AcO] , 94 -e [-H2 O] e 96

95 ------...

97 HO

NC R NC R

R I

94

95

Scheme 13: Synthesis of model ligands using the Baylis-Hillman reaction.

The purpose of acetylatmg the hydroxyl group in the Baylis-Hillman product is, of

- course, to provide a better leaving group, thus facilitating allylic displacement by

the nucleophilic 2° amine._One drawback o~ the acetylation approach, however, is

that the acetylated product can cyclise at elevated temperature to form an

Discussion 67

indolizine. 148 This cyclisation can be avoided by ensuring a reaction temperature

of less than ca. BO°C. The acetylated product was obtained in 60% yield, and was

readily identified by the characteristic acetate methyl signal at 02.21 ppm in the 1 H

NMR spectrum (Figure 24).

93

~~rrrl ~~I ~'T"~

PPII 10 9

r J JJ f

Figure 24: The 400 MHz 1 H NMR spectrum of the acetylated pr<!>duct 93 in CDCI3 · J

. The 1H NMR spectrum of the substitution product 96, on the other hand, (Figure

25) is characterised by the absence of the acetate methyl signal and the

appearance of a signal at 03.29 ppm corresponding to the allylic methylene

protons; the piperidine methylene protons are responsible for the signals in the

region 01.4-3.3 ppm. The presence of the nitrile group is confirmed by an IR

absorption band at 2217 cm-1. The analogous pyrrolidine derivative 97 was

similarly characterised trom the NMR and IR data.

96

J

2 ~ 0

r r j J __

Discussion

Figure 25: The 400 MHz 1H NMR spectrum of compound 96 in CDCI3 .

68

2.2.4.2 1,10-Phenanthroline-2,9-dicarboxaldehyde 65 as a Baylis-Hillman

substrate.

I Given the success obtained with pyridine-2-carbaldehyde 91, the use of 1,10-

phenanthroline-2,9-dicarbaldehyde 65 as a substrate for the Baylis-Hillman

reaction was investigated (see Scheme 12). Unfortunately, repeated attempts to

react the dicarbaldehyde 65 with acrylonitrile in the presence of DABCO, at room

temperature, in both chloroform and MeOH as solvent, proved unsuccessful. The

substrate 65 is insoluble in both solvents and, consequently, the reactions were

also attempted at reflux temperature in order to improve solubility. However, these

attempts were also uns~ccessful and this approach was abandoned.

An alternative strategy involving the Baylis-Hillman reaction was then explored

(Scheme 13). In this appr;ach, it was intended that the chelating arms of the ligand

Discussion 69

should be attached to a 1 ,3-diaminopropane spacer rather than a 1,1 O-phenan­

throline spacer. Unfortunately, attempted condensation of the acetylated 8aylis­

Hillman product 93 with 1,3-diaminopropane also proved unsuccessful, as

indicated by the 1 H NMR spectra of the reaction products isolated by

chromatography.

o OHC

91

~CN .. DABCO,

CHCI3,

4d

Q HO~-

NC

92

-P ~ I.

AcO N 0.5 h, reflux

NC

THF

98

Scheme 13: Attempte~ synthesis of the ligand 98. I

Discussion 70

2.2.5 Macrocycle Syntheses

Cyclic polydentate ligands, which exhibit conformational flexibility, have been

found to act as effective hosts in binuclear and polynuclear complexes,151-155 and

such complexes are potential targets as supramolecular catalysts. 156 Macrocycles

with large cavities, in particular, have received attention as inorganic and organic

anion and cation receptors, affording complexes which participate in biological

processes. 157-160

. Representations of a proposed macrocyclic system and possible dicopper

complexes are shown in Figure 26. It is apparent that two coordination possibilities

can occur, viz., tetradentate coordination with the copper atoms aligned "vertically"

(b) and bidentate coordination (the rema}ning coordination sites being filled by

solvent donors) in which the copper atoms are "horizontally" aligned (c).

Spacer Spacer Spacer

F 8 F.\ . , N N N-------Cu-------N N, .N

I I I I I '::CU cli' I J

N N NBN N/ "-N

8 8 I \ I •

N-N

Spacer Spacer Spacer

(a) (b) (c)

Figure 26: Representations of the proposed macrocyclic ligand (a) and possible dicopper coordination options, (b) and (c).

,-

Discussion 71

q:o H2N~NH2

~ 65 -* OHC CHO MeOH N N

"~X) 1 ( ) 100 I MeOH

HzN reflux N N

p=s qp N N a D 99 N N

~ Scheme 14: Synthetic approaches to macrocyclic ligands.

The synthesis of the polydentate macrocycles 99 and 100 was explored, the former

being more rigid and bulkier than the laUer. The intention was to compare the

catalytic activity of a rigid, bulky copper complex with that of a more flexible, less

bulky system. 1,1 0-Phenanthroline-2, 9-dicarbaldehyde 65 was used as a'spacer

and treated with 1 ,2-diaminobenzene and 1 ,2-diaminoethane with a view to

obtaining Schiff base macrocycles 99 and 100 respectively (Scheme 14).

The novel dibenzo analogue 99 was isolated in 30% yield and characterised by

MS and IR spectroscopy. The FAB mass spectrum clearly reveals the molecular

ion peak at m/z 616 (Figure 27), and high resolution MS analysis confirmed the

atomic composition. T~e IR spectrum has a strong band at 1619 cm-\

characteristic of the imine functionality. Interestingly, Jhe 13C NMR signals were

unusually broad, possibly due to conformational flexions which are intermediate

on the NMR time-scale.

Discussion 72

31!

136

49

197 309

41l

Iii

Figure 27: The FAB-MS spectrum of macrocycle 99 showing the molecular ion peak at m/z 616.

Unfortunately, attempts to obtain the known ligand161 100 were unsuccessful; the

products obtained in attempted preparations were insoluble in~ DMSO and are

believed to be polymeric. J

--

Discussion 73

2.2.6 Dendrimer Synthesis

In the last few years, much attention has been given to the preparation and

characterisation of dendrimer-based transition-metal complexes. 162 Such

complexes provide a promising structural concept for new materials. 163-168 Unlike

conventional metal-containing polymers, dendritic (from the Greek word meaning

branched) cascade molecules have the advantage of offering a highly controlled

architecture, which can act as a foundation for dendrimer -supported metal

complexes. 169 Research in this area has blossomed and is currently concerned

with exploring or developing various uses for dendritic molecules; these include:­

chemical sensors, magnetic resonance imaging agents, agents for delivering drugs

or genes into cells, micelle mimics, nanoscale catalysts, reaction vessels, immuno­

diagnostics, information-processing materjals and high-performance polymers.17o

In our investigation, it was antiCipated that dendrimer moieties could be used to

develop ligand systems which could mimic the macro characteristics of an enzymic

active site, viz., a hydrophilic exterior and a hydrophobic binding pocket. The

account which follows details our first endeavours in this direction. The

trihydroxyamine 105 was identified as a potentially useful dendri(l1er group, and

its synthesis, following Newkome's method,171 is outlined in Scheme 15. I

'-CN CN COzH

~CN HCI ~Co..H C~~ ,... ~ 103

Triton B 101 CN COzH

102 ! BH,.THF

~

~ T-1

~OH Raney Ni 105 H2 OH ~ 02~ OH 104

3 atm OH OH

Scheme 15: Synthesis of the dendrimer moiety 105.

Discussion 74

Nitromethane 101 was reacted with acrylonitrile in the presence of Triton 8 (a

surface active agent) to afford, via a Michael reaction, tris(2-cyanoethyl)­

nitromethane 102.171 The nitro triacid 103, prepared by acid-catalysed hydrolysis

of the trinitrile 102, was identified by the absence of a nitrile band (at ca. 2200

cm-1) and the presence of a carbonyl band (at 1721 cm-1) in the IR spectrum and

the appearance of a carbonyl carbon signal at 175.6 ppm in the 13C NMR

spectrum. The reduction of the triacid 103 to the trial 104 was achieved using the

mild reducing system, borane-THF;171 the absence of a carbonyl carbon signal and

the presence of a signal at 60.3 ppm (corresponding to CH20H) in the 13C NMR

spectrum served as evidence for the formation of the trial 104. The final reduction

of the nitro functionality to obtain the amino trial 105 was achieved by

hydrogenation using specially prepared T -1 Raney Nickel172 as a reducing agent.171

The success of this reduction was evidenced by the disappearance of the low-field

(94.0 ppm) C-N02 signal in the 13C NMR spectrum.

With the dendrimer moiety 105 in hand, attention was given to the construction of

dendrimer-based ligands. Several approaches were explored (Scheme 16). The

first route (Path 1) envisaged deprotonation of neocuproine 64, using NaH to

generate a dianion, for reaction with acrolein to produce the lialdehyde 106.

Subsequent reaction with the aminotriol 105 was then expected to afford the

dendrimer ligand 107 via a Schiff base reaction. Unfortunately, this synthetic

pathway was complicated by the formation of what appeared to be a polymeric

product (insoluble in DMSO) during the reaction with acrolein.

In further approaches (Paths 2 and 3), the tri- and tetracarbonyl derivatives 108

and 110 were expected to react with the amino trial 105 to afford the

corresponding dendrimer-based ligands 109 and 111, respectively. Numerous

attempts to oxidise the central ring in neocuproine 64, _using methods reported for

1,1 O-phenanthroline, 173-175 proved unsuccessful.

--

Discussion 75

106 107

a a

Path 1

Q KMnO" KOH -X-ii) Se02 Path 2 " Path 3 * i) HNQ"H,sO"KBr

ii) Se02

a a a

dt~ ~R~ 110 108

OHC CHO OHC CHO

1 RNH2 RNH21 \

I

NR a a

~ -N N- 111 109 ~

NR RN

~H R = OH

OH

Scheme 16: Attempted pathways to d~ndrimer-based ligands containing the 1,1 O-phenanthroline spacer.

Discussion 76

Finally, selenium dioxide oxidation of neocuproine' 64 afforded the dialdehyde 65,

which on treatment with the amino triol105, gave the dendrimer-based ligand 112

in 96% yield (Scheme 17). This ligand was characterised by IR, 1H and 13C NMR

spectroscopy. The 1H NMR spectrum (Figure 2S) is characterised by the imine

proton signal at oS.09 ppm and the broad upfield signals corresponding to the

dendrimer methylene groups. The FAB mass spectrum of ligand 112 did not exhibit

the expected molecular ion peak at m/z 610 but, rather, a peak at m/z 633

corresponding to attachment of a sodium cation (633.3627; M + Na+); such ions

are not unusual in the FAB-MS spectra.

64 o Q 65 OHC CHO

h'0H ~~OH

OH

112

Scheme 17: Synthesis of the dendrimer -based ligand 112.

I

A mononuclear copper complex of this ligand was considered unlikely to be a good

biomimetic model since two copper ions are found in the active site of the

-. - tyrosinase enzyme, whereas the coordination cavity in-ligand 112 would probably

only accommodate a single copper ion. Complexation with a metal would, however,

establish whether coordination involves the aromatic nitrogen atoms or whether the

Discussion

hydroxyl groups of the dendrimer moieties compete as donors.

pp.

112 ~

~:JfN ~OH HO OH OH

I 10

77

. :

\ Figure 28: The 400 MHz 1H NMR spectrum of the dendrimer-based ligand 112.

I

Discussion 78

2.3 COMPLEXATION AND COMPUTER MODELLING STUDIES

2.3.1 Complexation Studies

While the ligands synthesised in this study were designed to form dinuclear copper

complexes, they may also coordinate with metals other than copper, e.g.

manganese(II), iron(II), cabalt(ll), nickel(II), platinum(II), palladium(ll) and lead(II).

Consequently, some attention was also given to the complexation of ligands

containing the biphenyl and 1,1 O-phenanthroline spacers (Schemes 2 and 8) with

cobalt(II), nickel(II), platinum(lI) and paliadium(II).

Hard metals, which include copper(II), cobalt(ll) and nickel(lI) are known to

coordinate with amide oxygen,176-178 but the resulting metal-oxygen bonds are

weaker than the corresponding metal-amide nitrogen bonds. Coordination of

metals to nitrogen donors occurs over a wide pH range, but the ability of metals to

deprotonate and complex with amide nitrogen is pH-dependent. From

potentiometric titration studies of divalent metal ion-dipeptide systems, the

following pH preferences have been obtained for the coordination of metals to

peptide nitrogen donors:- copper(II), pH 5 - 6;179.180 nickel(II), RH 8 _ 9;181.182

platinum(II), pH < 5; 183 and paliadium(II), pH 2.5 _ 3.5.184.185

It must, however, be emphasised that in the present complexation studies, the pH

was not tuned to specifically achieve coordination of the metal with the amide

nitrogen donors, and coordination of the metal could involve either the oxygen or

nitrogen atom of the amide functionality.

The analysis of these complexes was expected to reveal:­

(i) whether dinuclear complexes were formed;

-. - (ii) whether coordination of the metal occurs through-the. nitrogen or oxygen

atom of the amide Ijgsnds; and

(iii) the preferred geometry of the resulting complexes.

Discussion 79

Ligands containing amide, imidazole and benzimidazole moieties should provide

coordination environments that are similar to those in biological systems, such as

enzymes, and the resulting complexes of cobalt and nickel could prove to be good

synthetic models of metalloproteins in which these metals are found. The

structures proposed for the various complexes, which have been isolated, are

based on a consideration of the IR, NMR, UV-Vis, microanalytical and computer

modelling data. Unfortunately in only two cases were crystals obtained which were

suitable for X-ray crystallographic analysis.

2.3.2 Copper Complexes

Copper is present at low concentrations in many enzymes and proteins. 18G In larger

quantities, copper is toxic to humans and its salts are highly toxic to lower

organisms. 187 Copper ions instantly form complexes with a wide variety of ligands

and exhibit a range of oxidation states in the coordination compounds in which

they occur. The copper(l) and (II) oxidation states are the most prevalent, but the

copper(lll) and (IV) oxidation states are also known. Copper(III) and copper(IV)

oxidation states are, however, rare. Copper(l) is readily oxidised to copper(II),

which is the most stable oxidation state of copper under standard cpnditions. l88, 189

A wealth of copper complexes have been generated,189and the crystal stgJctures

of many of these have been deterrT)ined. These structures have permitted the

characterisation of various regular and distorted geometries of the copper(II)

ion, 190,191 which are associated with the Jahn-Teller effect. 192 Copper(I), a d10 ion,

behaves as a soft acid and favours coordination to soft sulfur and phosphorus

bases,193whereas copper(II), a d9 ion, is considered to be a border-line hard acid

which favours oxygen and nitrogen donors.l94 Diamagnetic, colourless complexes

are formed by copper(l) with its closed shell configuration, 190,195 while copper(II)

produces highly coloured paramagnetic complexes. 190,194, 195

Discussion 80

2.3.2.1 Synthesis of Copper complexes

The complexation of copper with the various ligands occurred with ease [see

Experimental section (3.3.1 )]. Depending on the solubility of the ligand, dry DMF

or dry MeCN were used as solvents in these complexation reactions.

Biphenyl and 1,10-Phenanthroline Complexes

The biphenyl (Scheme 18) and 1,1 O-phenanthroline (Scheme 19) ligands were

reacted with [Cu(MeCN)4]PFs, a Cu(l) reagent, at room temperature under N2 in

dry, degassed solvent; in some cases, the solutions had to be heated to dissolve

the ligands. The ligand solution was added dropwise to the solution of the

copper(l) reagent to obtain dinuclear copper(l) complexes. The biphenyl

complexes, obtained using this procedur~, were green, indicating the copper(lI)

oxidation state. The 1,1 O-phenanthroline complexes, however, were brown in

colour and there was some uncertainty as to whether this was an indication of the

copper(l) or (II) oxidation state, since colourless complexes are formed for

copper(l) and green complexes are formed for copper(II). Microanalysis data for

these complexes (Table 8, p.85) show that the biphenyl and 1,1 O-phenanthroline

complexes are, typically, dinuclear and of acceptable purity; the d~ta for complex

117 appears to be a mixture of mononuclear and dinuclear complexes (9ata not

included in Table 8). The results also ~ndicate that two PFs- anions are present as

counter ions- except for complex 119 which has four. This suggests either that

protons from the amide, benzimidazole or imidazole moieties are generally

removed during complexation, or that both copper atoms are in a copper(l)

oxidation state in these complexes. All the complexes, except complex 119, were

insoluble in DMSO due, it is believed, to the formation of polymers. The structures

proposed for the copper complexes detailed in Schemes 18 and 19, while

necessarily tentative, are-based on a consideration of the spectroscopic (Sections

-. - 2.3.2.2-4) and microanalytical data.

Discussion

+ a b C ~N

---+--+-±-+ ~N~ 0 -

R N~; HN "N ; r-(~ I N ~ b V

~ b TN - \ ~ b H N

f' N~h ~U(MeCNZ4PF6 - 0 \ ./"N~

DMF CU

0' [for 55b] QN/ \ R ...............

HN ~ 0 ~ /I N~ DI'vF

o ~ NH~R + I~

55a-c

MeCN * [for 55a] [Cu(MeCN)4lPF 6

117

yU~N)4PF6 DMF ~for 55e]

115

Scheme 18: Complexation of the biphenyl amide ligands with copper.

81

Discussion 82

a b c

69a-c

ICU(MeCNl4II'FSl OMF

118a-c

I

68a 119

Scheme 19: Complexation of selected 1,1 O-phenanthroline ligands with copper.

Schiff base complexes

It was felt that it would be better to react the Schiff base ligands with a copper(lI)

reagent as the formation of mixtures of copper(l) or ~II) complexes might have

caused complications during the isolation and purification of the biphenyl- and the <

1,10-phenanthroline-based complexes. Consequently, the ligands were treated

with CU(N03)2.3H20 in MeCN (Scheme 20), without the precautions used

Discussion 83

previously for the copper(J) complexes, viz., pre-drying and degassing the solvent

and conducting the reaction under N2. The resulting complexes 120, 121 and 123

were green in colour and tended to be hygroscopic; from their microanalysis data

(Table 8), it can be seen that these complexes were dinuclear. However, the data

obtained for the complexes with ligands 74, 75, 78 and 84 (not included in Table

8) suggest formation of mixtures of mononuclear and dinuclear complexes. While

a structure has been proposed for complex 120 (Scheme 20), the complexity of the

spectroscopic data for complexes 121 and 123 preclude the confident assignment

of structures.

:l4+ YI( H~ YI( OH2 4N03

X,Z \ / 76 fNH N'l .. Cu fNH Nt Cu

MeCN H~-- \ / "-Py Py ~ Py Py O~

R 120

yY X,Z fNH N'l .. 121 123

MeCN R Me Benzyl

Py Py

77 79

R Me Benzyl J

Py = © x = Py ~N~ Z = CU(N0:Y2' 3 H2O

Scheme 20: Complexation of the Schiff base ligands with copper.

The Macrocyclic complex

When a solution of the macrocycle 99 in DMF was added dropwise to a solution

of the copper(lI) reagent Cu(N03)z.3H20, the coloL!r of the reaction .mixture

changed from turquoise to red-brown. Microanalysis of the resulting complex 124

(Table 8) indicated it to be dinuclear and "of acceptable purity and, due to the

presence of four nitrate anions, to contain two copper(lI) cations.

Discussion

p=ts N N a 0 DMF

N N

Q9 99

R814+

f_~ :_' I 4N~ \ I

.---Cu-.. N N a 0 N-.. __ N

Cu / \

-N N-

124

84

Scheme 21: Complexation of macrocycle 99 with copper.

Table 8: Micro-analytical data for the copper complexes followed, in parentheses, by the calculated values.

Complex Complex

Stoich iometry

%Carbon %Hydrogen %Nitrogen

---------------------------------------------------------------------------------------

115 Cu2( 55b )(PF 6h.2DMF 41.9 (41.9) 3.9 (3. 7) ~ 10.0 (10.3)

116 Cu2(55c)(PF6h·5Hp 30.8 (30.9) 3A (3.5) 8.8 (9.0) J

118a Cu2(69a)(PF6h· 5Hp 37.7 (37.3) 2.0 (2.7) 9.7 (9A)

118b Cu2(69b )(PF 6h 6H2O 35A (35.6) 3.2 (3.6) 10.3 (10A)

118c Cu2(69c)(PF6h·5Hp 29.5 (30.0) 3.0 (3.2) 12.3 (11.7)

119 Cu2(69a)(PF6)4 30.2 (29.2) 2.7 (2.3) 7A (7.3)

120 Cu2(76)( N03)4' 7H2O 28.2 (28.2) 3.9 (4.7) -13.7 (13.8)

121 Cu2(77)(N03)4· 8Hp 29.0 (28.6) 4.1 (5.0) 13.0 (13.3)

123 Cu2(79)(N03 )4· 9Hp 32.8 (33A) 3.6 (5.2) 12.0 (12.0)

124 Cu2(99)(N03)4' H2O 47.5 (47.6) 3.0 (2.6) 16.6 (16.7)

Discussion 85

2.3.2.2 NMR Studies of the copper complexes

Upon coordination of copper the ligand nuclei are deshielded relative to the free

ligand, and the magnitude of the downfield shift may reflect, to some extent, the

proximity of a specific nucleus to the coordinated copper cation. 92 While copper(l)

complexes can be readily analysed by NMR spectroscopy, copper(lI) complexes

are paramagnetic, and the signals for the nuclei close to the copper tend to be

broadened, thus complicating analysis. This difference provides a quick and

simple distinction between copper(l) and copper(lI) complexes.

It was not possible to distinguish, on the basis of colour, whether copper(l) or (II)

1,1 O-phenanthroline complexes were obtained. From the 1H NMR spectra of these

complexes, however, it could be seen that the copper(l) oxidation state was not

dominant as the signals in the 1 H NMR spectra were broad. Thus, it seems, that

at least one copper atom in the 1,1 O-phenanthroline copper complexes may be in

a copper(lI) oxidation state; magnetic susceptibility studies- will be required to

determine whether copper(l) is present in these complexes.

2.3.2.3 IR studies of the copper complexes

J

These studies were undertaken to determine whether coordination of the ligands

with copper occurs through the nitrogen or oxygen of the amide functionality, or

through the secondary or tertiary amine groups of the imidazole and benzimidazole

units. The spectra were also expected to show whether the amide proton is

removed during complexation. It was, however, not possible to establish whether

coordination involves the 1,1 O-phenanthroline nitrogen donors because the

aromatic bands overlap in the region of interest (1600-1500 cm-1).

- Biphenyl- and 1, 10-Phenanthroline-based complexes

It can be seen from the €lata in Table 9 that copper coordinates with either the

amide oxygen or nitrogen in the 1,1 O-phenanthroline- and biphenyl-based copper

complexes. A negative carbonyl shift indicates coordination through the carbonyl

Discussion 86

oxygen, while a positive NH shift indicates coordination through the nitrogen of the

amide group. Thus, for complexes 118a and 118b, coordination involves the amide

oxygen and, due to the absence of an amide NH band, through the amide nitrogen.

For complexes 115 and 116 coordination appears to involve the amide nitrogen

due to the positive NH shift. In the case of complex 118c, coordination also

involves the amide nitrogen since the amide NH band is absent indicating that

deprotonation of the amide NH had occurred upon complexation with copper; the

fact that f1vc=o is zero, however, casts some doubt on the carbonyl oxygen

coordination depicted in the structure proposed for complex 118c (Scheme 19).

For the PFe- anion a P-F stretching band between 840-850 cm-1 is observed for

each of these copper complexes. 196

Table 9: Selected IR data for the amide copper complexes of the biphenyl- and 1,10-phenanthroline-based ligands .

Copper complexes .6V a/cm -1 NH .6Vc=Oa/cm -1 VPFs

b/cm-1

115 65 33 845

116 55 29 846

118a - -21 842

118b - -13 ~ 846

118c - 0 845 I

a Frequency shift for amide apsorption on complexation. A positive value indicates a frequency increment and a negative value a decrement relative to the free ligand.

b P-F stretching band for the PFe- counterion.

In the spectra of the biphenyl complex 115 and the 1,1 O-phenanthroline complex

118b, the benzimidazole NH band (at ca. 3171 cm-1) is absent, indicating that this

proton has been removed suggesting that coordination with copper occurs through

the secondary nitrogen -<:Ionor of benzimidazole. The fact that the amide protons

_ are not removed on formation of complex 115, suggests that the pKa of the

benzimidazole protons in the ligand 55b is lower than that of the pKa of the amide ~ ,;a.. ;:

protons. In the IR spectrum of the biphenyl complex 116 the imidazole NH band

is present (at 3156 cm-1), but the frequency of the amide NH band is increased by

Discussion 87

55 cm-1 to 3242 cm-1 on complexation. The presence of the imidazole NH is an

indication that the tertiary imidazole nitrogens as well as the amide nitrogens (see

above) are, coordinated to copper. For complex 116, it seems that the reaction

conditions do not favour deprotonation of either the imidazole secondary amine or

the amide groups .

.,. 1248.2

" 1216.9

1361.9 /

" \

1492.!!-

\il ...1.\

1;2'92.2

J Figure 29: IR spectrum of complex 115 in hexachlorobutadiene.

A band at 3273 cm-1 in the IR spectrum of the bis(amino) 1,1 O-phenanthroline

complex 119 is attributed to the secondary amine NH stretch, indicatfng that the

amine protons are not removed upon coordination with copper. A band at 853 cm-1

is observed for the PF 6' anions in this complex.

Schiff base complexes

The IR spectra of the copper complexes prepared from the Schiff base ligands are

more complicated than expected. This complexity is attributed to the presence of

the nitrate anion, which can act in its own right as an un identate or bidentate ligand

or simply be present as an uncoordinated counterion.202 All these possibilities have

Discussion 88

been found to occur in the Schiff base copper complexes.

The complex 121 exhibits a number of significant bands (Figure 30). These

include:-

i) the imine C=N band at 1637 cm-1;

ii) bands at 1480 and 1283 cm-1 characteristic of the nitrate anion coordinated

as a bidentate ligand;

iii) a band at 1445 cm-1 indicating un identate coordination of nitrate; and

iv) a band at 1013 cm-1 which suggests that the nitrate anion is coordinated as a

un identate ligand;197 a second band expected at ca. 1315 cm-1may be

masked by the band occurring at 1383 cm-1 and indicates the presence of

free nitrate ion.

It is concluded that, in addition to the presence of the nitrate ion, unidentate and

bidentate coordination of the nitrate ion occurs in this copper complex.

'" "

v1'.\ r ,,;,., \

"".' \ 804 8 " "

2926.6 IJ\ /;" \ '/ \

"T • I

71'0.1

; \

JOn.3 29-12.9

3413.1

\ \ " / \\ \\"''-'" 1292.8

1653.0 "-

I, lO!J.5 1393.1 11;31.5 \

I i 1444.9

'. ·1

... 1606.7 1560.4

Figure 30: IR spectrum of complex 121 in KBr.

Discussion 89

The IR spectrum of the copper complex 123 exhibits bands corresponding to the

imine moiety and the free and bis-coordinated nitrate anion. In the case of the

macrocyclic complex 124, however, no un identate or bidentate coordination of the

nitrate anion is apparent; the band at 1383 cm-1 reveals the presence of free nitrate

anions, while the imine C=N band appears at 1653 cm-1. All of the Schiff base

copper complexes examined proved to be hygroscopic and, consequently, the

assignment of the NH band is precluded by the presence of a broad H20 hydroxyl

band.

2.3.2.4 UV-Vis studies of the copper complexes

It was decided to use UV-Vis spectroscopy to detect whether the two copper atoms

are in the same or different coordination environments within the dinuclear

complexes. Different geometrical arrangements of the copper atom give rise to

different absorption characteristics and, as the copper(l) and (II) oxidation states

favour different geometries, it was expected that the UV-Visible absorption data

would reflect the oxidation state(s) involved. Due to the insolubility of the

complexes in DMF, DMSOwas used as solvent but, unfortunately, the regions of

interest were masked by solvent absorption bands, thus precludi~g comparison

with the UV data reported for the enzyme, tyrosinase. J

Discussion 90

2.3.3 Computer Modelling Studies

Molecular mechanics is now a routine tool in organic chemistry,198,199 and has

become increasingly important in the field of drug design200 and in studying the

matching of bases in nucleic acids. 201 Its use is also well established in

coordination chemistry,202,203 where it has been used for the analysis of disordered

structures,204 the prediction of unknown structures,205 the determination of isomer

and conformer ratios206 and metal ion selectivities. 207

Molecular modelling and molecular mechanics

The basis of molecular modelling is that all the important molecular properties (i.e.

stabilities, reactivities and electronic properties) are related to the molecular

structure. A computational study can only be carried out once a molecular model

has been established, and because of its computational simplicity and efficiency, -

molecular mechanics is commonly used to construct such models. Typically, the

energy-minimised model, which is obtained, represents an idealised gas-phase

structure of the molecule that is independent of solvent effects in solution, or of

lattice interactions in the solid state. I

The molecular modelling of transition' metal complexes is complicated, however,

by the presence of partially filled, metal ion d-orbitals, which are responsible for

the multifarious structures of coordination compounds, which exhibit a ,variety of

coordination numbers and geometries. Moreover, metal-ligand bonds can become

elongated during steric crowding, especially when small metal ions are

coordinated. In cases where the ideal bond lengths have not been determined, it

is considered acceptable to use an average force constant. 203

In the computer models illustrated in this section, the following colour key applies:

carbon: light-blue; nitrogen: dark-blue; oxygen: red; hydrogen: white; copper

yellow.

Discussion 91

2.3.3.1 Computer modelling of the dinuclear copper complexes

Computer modelling studies were undertaken to explore the structures of proposed

dinuclear copper complexes and their potential as biomimetic tyrosinase

analogues. Bernhardt and Comba have published a force field for copper(lI) in

organometallic complexes. 62 This has been determined by modelling a series of

amine and imine complexes and correlating the results with X-ray crystal

structures. A close similarity was observed for these correlations, thus providing

support for the validity of computer modelling of copper complexes. The discussion

that follows focusses on the modelling of possible monomeric structures for the

dinuclear copper complexes 120, 128 and 130t. The aim, in each case, was to

determine:- (i) the most likely conformation for the ligand and the complex; (ii) the

Cu-Cu distance in the complex; (iii) the fe~sibility of a dioxygen peroxide bridge

between the two copper atoms; and (iv) the accessibility of the Cu-02-Cu unit for

binding phenolic substrates.

The distance between the copper atoms is, of course, crucial since the initial

stage of the enzyme-catalysed reaction involves the binding of dioxygen across

the dinuclear copper site. For rigid, planar molecules, a Cu-Cu distance of 5 A is

required for a good steric match; less for more flexible systems. 20B lt is beneved

that the Cu-Cu distance in the tyrosinase enzyme is 3.5 A,32,33 and models

reported to have displayed successful biomimetic activity have a similar Cu-Cu

separation. 101 ,217 Initially, it was thought that the dioxygen in tyrosinase was bridged

in a cis fJ-1,2 fashion209 but, in fact, it is coordinated in a fJ-rj-:rj- side-on bridging

mode.210,211 The binding of the substrate appears to be accompanied by a change

in coordination from tetragonal copper(l) to trigonal bipyramidal copper(II).212 It is

also believed that the peroxide moiety is in an equatorial plane when the phenolic

substrate binds axially to one copper atom in the active site.213 However, in order

for the electron transfer to occur most readily at the active site of tyrosinase, the

t While the apparent polymeric nature of the copper complexes has already been indicated (p.81 and p.82) it is possible that, in solution, catalytically active monomeric complexes may exist in equilibrium with polymeric complexes.

Discussion 92

phenolic substrate has to be in an equatorial plane to permit maximum overlap of

the substrate donor orbitals with the half empty d/./ orbitals of the copper(ll)

ions.211 Such rearrangement from axial to equatorial binding is not required for

catechols and, as a result, their oxidation is less geometrically and sterically

demanding. Apparently, the protein pocket contributes to the stabilisation of the

binding of either substrate through n-stacking interactions.213

In the present study, two modes of binding for dioxygen were modelled for each of

the copper complexes 120, 128 and 130, i.e. a fJ -1,2 end-on and a fJ-r1:r1 side­

on bridging mode since these have been reported in the literature for copper

complexes. 214.215.216

Energy-minimised structures for the ligands, their dicopper complexes and oxygen­

bridged derivatives were modelled using MSI Cerius2 software on a Silicon­

Graphics 0 2 platform. Solvent molecules can coordinate to copper and, as a result,

the binding of the solvent, acetonitrile, was taken into account during the modelling

of these complexes. Pertinent data for the complexes are summarised in Tables

10 and 11. The ligands so examined (56b, 68b and 76) are depicted below.

Table 10: Cu-Cu separation in the modelled complexes before dioxygen binding.

Ligand Complex Cu-Cu separation (A)

76 120m 6.836

56b 128m 8.451

68b 130m 4.789

a m = monomeric form

Discussion

Table 11: Cu-Cu separation and the potential energy in the modelled complexes after dioxygen binding.

Ligand Complex' Cu-Cu separation (A) Potential Energy (kcalmol-1)

(after 02 binding)

56b 1290s

56b 1290e

68b 1310s

68b 1310e

76 1320s

76 1320e

as = J.l-ti:ti side-on dloxygen bndgmg; oe = J.l-1,2 end-on dioxygen bridging.

4.880

1.756

2.188

4.616

2.117

4.642

-246.9

644.9

-38.2

1030.3

-687.8

48.0

J

93

Discussion

Complexes containing the biphenyl spacer

(a)

(b)

Figure 31: Energy-minimised conformations of the ligand 5Gb (a) before and (b) after complexation with copper.

94

From the models in Figure 31 it can be seen that there is not a drastic change in

conformation when ligand 5Gb coordinates with copper. Upon binding to dioxygen,

the copper atoms are, of course, brought much closer together. The models for the

fJ-rT:rt side-on (complex 1290s) and fJ-1,2-trans bridging (complex 1290e) of

dioxygen are shown in Figure 32 (a) and (b) respectively. The fJ-1,2-trans bridging

mode is expected to be favoured since this is the more stable arrangement for

dioxygen in dinuclear copper complexes.21o Complex 1290e has a Cu-Cu distance

of 1.756 A, but it is less stable than complex 1290s, which has a Cu-Cu distance

of 4.880 A. It is apparent that there is a considerable change in conformation when

dioxygen is bound to the dinuclear complex 128, a.nd that the geometry of the

metal atoms changes from tetrahedral to octahedral in complex 1290s, and to

trigonal bipyramidal in-complex 1290e.

Discussion

(a)

(b)

J

Figure 32: Energy-minimised conformations of the dioxygen-bridged copper complex 128, showing (a) side-on and (b) trans bridging.

95

Discussion 96

(a)

I

Figure 33: The space-filling models of the energy-minimised conformations of the dioxygen-bridged copper complex 128, showing (a) side-on and (b) trans bridging.

In the dioxygenated sysfem, the biphenyl arene rings remain out of the plane of the -

copper atoms and the binding site is seen to be concave. From the space filling

models of complexes 1-2905 and 1290e (Figure 33), it appears that, in both cases,

the active site is readily accessible.

Discussion 97

Reglier has reported a similar model of a complex that also features the flexible

biphenyl spacer (see section 1.3.5 p.32) and displays both phenolase and

catecholase activity when dioxygen is bound in the I-l-oxo form. 103 It has been

shown, by means of molecular mechanics calculations using a BIOGROMOS

program, that this model reflects a Cu-Cu distance of 3.6 A when the deoxy form

exhibits a biphenyl dihedral angle of 60°.217 In the light of these results it is

expected that the complex 128 should also be capable of attaining a Cu-Cu

distance of ca. 3.6 A and that it could also display phenolase and catecholase

activity.

Complexes containing the 1,10-phenanthroline spacer

In the model of the dicopper deoxy complex 130 the tetrahedral geometry of

copper(l) is apparent. Upon oxygenation, the geometry changes to a distorted -

octahedron in the case of the fJ-r/: r/ side-on bridged complex 13105; in the fJ-1 ,2-

trans complex 1310e (Figure 34), however, the tetrahedral geometry changes to

square pyramidal at the one coordination site and trigonal bipyramidal at the other.

The Cu-Cu distance in complex 13105 is 2.188 A, and 4.616 A in cbmplex 131oe. J

From the potential energies it appears that complex 13105 is more stable than

1310e, but in neither case does the oxygenated complex provide an obvious

binding pocket to accommodate the substrate.

Discussion 98

(a)

(b)

I

Figure 34: Energy-minimised conformations of the dioxygen-bridged copper complex 131 oe: (a) a cylinder model and (b) a space-filling model.

Discussion 99

In summary, the biphenyl spacer, due to its flexibility, should allow the oxygenated

complex (12905 or 1290e) to adopt an optimal conformation that permits

coordination of the substrate between the two copper ions. This may, however, be

more difficult in the 1,1 O-phenanthroline complexes 13105 and 1310e because of

the rigidity imposed by the 1,1 O-phenanthroline spacer. In the latter complexes,

the "active site" is peripheral, the dioxygen bridge being situated above the plane

of the complex. The biphenyl analogues (12905 and 1290e), however, exhibit a

concave region about the "active site" which could act as a binding pocket,

providing access to a substrate molecule.

Schiff base complexes

The acyclic Schiff base ligands were expected to form complexes which would be

considerably more conformationally flexible than the biphenyl or 1,10-

phenanthroline analogues. For ligand 76, a change in conformation occurs upon

coordination with copper and tetrahedral geometry is observed for the copper

atoms in the model of the resulting complex 120. This tetrahedral geometry

changes to distorted octahedral in the case of the l1-rf:rf side-on bAdged complex

13205 and, in complex 132oe, where dioxygen is bound in a 11-1 ,2-trans Bridging

mode, the geometry changes to trigonal bipyramidal (Figure 52) . The Cu-Cu

distance is 2.117 A in complex 1320s.and 4.642 A in complex 1320e. Comparison

of the potential energies of these complexes reveal that complex 13205 is more

stable. From the models of the complexes, it can be seen that, in each case, the

active site lies in a concave region constituting a convenient binding site readily

accessible to substrate molecules.

Discussion 100

(a)

(b)

I

Figure 35: Energy-minimised mod~ls of (a) complex 13205; and (b) complex 1320e.

Unfortunately, none of the complexes afforded crystals suitable for single crystal

X-ray analysis, thus precluding direct comparison of actual and computer-modelled

structures. Moreover, in the case of the biphenyl- and 1,1 O-phenanthroline-based

systems, the formation of polymeric complexes was observed.

Discussion 101

2.3.4 Other Transition Metal Complexes

2.3.4.1 Cobalt Complexes

Cobalt may exist in various oxidation states, viz., cobalt(I), cobalt(II), cobalt(lIl) and

cobalt(IV).218 The cobalt(l) ion with its eight d electrons occurs in many complexes

with n -bonded ligands. To make these complexes, in which the metal is trigonal

bipyramidal or tetrahedral, CoCI2 is reduced with Zn or N2H4 in the presence of the

ligand. 219 The reduced form of Vitamin B12 is known to contain Co(l).220 The

cobalt(lI) ion has a d 7 configuration and typically exhibits four-coordinate

tetrahedral or six-coordinate octahedral stereochemistry. A small difference in

stability exists between these arrangements and, therefore, they may occur in

equilibrium.

Cobalt(III), a d 6 low spin, diamagnetic cation, is kinetically inert and displays

octahedral stereochemistry. This is the most common oxidation state for cobalt in

complexes, and cobalt(lI) complexes are readily oxidised to cobalt(III). The

oxidation of cobalt(lI) to cobalt(llI) is attributed to the higher crystal field

stabilisation energy of cobalt(III).221 Both cobalt(lI) and (III) pre\er nitrogen or

oxygen donor ligands. The cobalt(lV) oxidation state is the highest oxidatign state

known for cobalt, but this is not as qommon as the cobalt(lI) and (III) oxidation

states.

Cobalt plays an important role in biological systems such as enzymes. The

enzyme, .glutamic mutase, is involved in the metabolism of amino acids, and

ribonucleotide reductase in the biosynthesis of DNA221 Vitamin B12, which was the

first naturally occurring organometallic compound to be identified, is a cobalt

complex which is involved in alkylation reactions. 218,222,225 In biological systems,

- - iron(ll) (e.g. haemoglobin and myoglobin) or copper(IHhaemocyanin) complexes

serve as reversible oxygeo carriers for the t~ansport and storage of oxygen. 223, 224

Since the discovery that Co(lI) forms simple complexes that react reversibly with

oxygen to give 1: 1 and 2: 1 metal complex-oxygen adducts, 225, 226 cobalt complexes

Discussion 102

have been widely studied as models for biological processes. 15, 227,228 Tetradentate

Schiff base cobalt complexes have been used to study dioxygen activity and, since

they undergo alkylation reactions, they have been used to mimic the reactions of

Vitamin 812.

2.3.4.1.1 Synthesis of cobalt(lI) complexes

Cobaltous chloride hexahydrate (CoCI2.6H20) was used to prepare the cobalt

complexes, following the procedures detailed in the Experimental section (3.3.2).

The microanalysis data for the resulting complexes, given in Table 10, indicate that

four chloride anions are generally present and suggest that the amide, imidazole

and benzimidazole protons are not removed during complexation with cob.alt. The

complexation of the biphenyl and 1,1 O-phenanthroline ligands with cobalt are

outlined in Schemes 22 and 23. The formation of complex 135 (Scheme 22) was

unsuccessful as could be seen from the microanalysis data which are not included

in Table 12.

Table 12: Microanalytical data for the cobalt(ll) complexes followed, in parentheses, by the calculated values.

J

Complex Complex

Stoichiometry

%Carbon %Hydrogen %Nitrogen

----------------------------------------------------------------------------------------

133 Co2(55a)CI4·Hp 46.3 (46.3) 3.7 (4.3) 7.5 (7.7)

134 Co2(55b)CI4·8Hp 41.5 (41.3) 4.0 (4.8) 9.5 (9.0)

136 Co(61a)CI2 57.8 (58.7) 4.5 (4.7) 6.6 (6.5)

137a Co2( 69a )CI4 . 6HzO 39.1 (39.8) 3.3 (4.3) 10.3 (10.0)

137b Co2(69b )CI4. 8H2O 39.7 (40.2) 3.6 (4.4) 12.1 (11.7)

137c Co2(69c )CI4·2H2O 38.1 (38.4) 3.6 (3.5) 14.8 (14.9)

138 Co2(68a)614·8H2O 41.8 (41.3) 3.8 (4.9) 10.1 (103)

Discussion

55a-c

CoCI2. ~~;P / DMF,Me~

MeOH DMF

CoCb.6H;P * MeCN, MeOH

I abc

MeCN

61a

CI CI ,/

/Co'-...,. I

~N ~ I u N

H I

103

H N~

o jN~ ~c/ C( "CI 133

134

135

136 I

Scheme 22: Complexation reactions of the biphenyl ligands with cobalt(II).

Discussion

R

~ p=q DMF Ij ~ Ij ~ a 0 COCI2.6H;P -N N-

Ib

o 0 .. HNtN HN 0 0 NH NH HN r C~-CI CI_\ \

b ) 2 0 I o-R R R

CI \ CI tN 70a-c 137a-c

c

I H~

M-N~ IjN-~ MeCN

COCI2.6Hp

or

68a 138a 138b

or

I

138c

Scheme 23: Complexation reactions of the 1,1 O-phenanthroline ligands with cobalt(II).

104

Discussion 105

2.3.4.1.2 Spectroscopic analysis of the cobalt(lI) complexes

1H and 13C NMR spectra of the cobalt complexes were run to define the oxidation

state of the metal ions. As cobalt can form both paramagnetic [Co(ll)] and

diamagnetic [Co(lll)] complexes, 1H NMR spectroscopy was expected to establish

the oxidation state of cobalt in the complexes. The 1H NMR spectra obtained for

the biphenyl- and 1, 10-phenanthroline complexes reveal broad, poorly resolved

signals, indicating that at least one of the cobalt ions in each complex is

paramagnetic.

IR spectra of the complexes were expected to establish whether the cobalt ions

coordinate through the amide nitrogen or oxygen atoms. Determining whether

coordination occurs through the aromatic. nitrogen donors, however, was not

possible, due to the congestion of bands in the aromatic region (1600-1500 cm-1)

which makes detecting a band shift in this region rather difficult. Spectra were run

in the far-IR region to determine whether the coordinated chlor-ide anions are in a

tetrahedral or octahedral environment. The IR frequencies for selected bands are

shown in Table 13. From the negative IR frequency shifts of the amide carbonyl

(amide I) bands, it is apparent that coordination occurs through the ~mide oxygen

atom in both the biphenyl and 1,1 O-phenanthroline cobalt complexes. Althol!9h the

benzimidazole NH band cannot be ider)tified in the IR spectra of the biphenyl and

1,10-phenanthroline complexes 134 and 137b respectively, it is assumed that

coordination does not involve the secondary benzimidazole nitrogen as the

microanalysis data show the presence of four chloride ions in each case. This is

an indication that no deprotonation occurs on formation of these complexes, and

it is believed that coordination involves the tertiary, benzimidazole nitrogen in both

of these complexes. For the 1, 10-phenanthroline complex 137c, the imidazole NH

bands are shifted, relative to the free ligand, indicating coordination through the

-. -tertiary, imidazole nitrogen.

Discussion

Table 13: Summary of the IR frequencies (v M-C') and the amide frequency shifts (/1 V NH and /1 V c=o) on formation of the amide cobalt complexes.

Cobalt Complex lw NH /cm·1 L1v c=o/cm-1 vM-c,/cm-1

133 - -19 306

134 13 -30 295

137a 18 -7 303

137b -62 -7 311

137c -38 -6 299

106

While the amide NH band cannot be seen in the IR spectrum of the biphenyl

complex 133, due to the presence of a strong, broad water band, the microanalysis

data clearly indicates that formation of the complex does not involve deprotonation.

An amide NH band is, however, observed ~,at 3274 cm-1 in the IR spectrum of the

1,10-phenanthroline complex 137a, thus confirming the microanalysis data

obtained for this complex which, due to the presence of four chloride ions, indicate

that the amide functionality is not deprotonated.

Two Co-CI bands characteristic of tetrahedral geometry are anticipated in the far­

IR region (ca. 301 and ca. 324 cm-1).229 These bands are not observed for these

~

amide complexes instead, a very strong, broad band, often with shouldJers, is

observed at ca. 300 cm-1. This may be due to accidental degeneracy of the

symmetric and antisymmetric Co-CI stretches, and it is suggested that the band at

ca. 300 cm-1 indicates a distorted tetrahedral cobalt geometry occurring within

these complexes.

Coordination of the ligand with the tetrahedral cobalt ion in the amine complex 136

was unambiguously established by single crystal X-ray crystallography of this

complex (Figure 36). A far-IR band at 298 cm-1 is consistent with coordination of

_the chloride ion in a tetrahedral arrangement.

Discussion

J

Figure 36: X-ray crystal structure of cobalt(lI) complex 136, showing the crystallographic numbering. "

107

Discussion 108

An IR band at 3398 cm-1 for complex 138 suggests that deprotonation of the

secondary amine does not occur during complexation, while a shift of 98 cm-1 for

this NH band (relative to the free ligand) reflects coordination to the aliphatic

amine nitrogen. In the far-IR region the strong, irregular band at 311 cm-1 suggests

that the chloride ions coordinate to cobalt in a tetrahedral arrangement.

Cobalt ions in tetrahedral or octahedral coordination environments give rise to

characteristic absorption bands in the visible region. Absorption data for the cobalt

complexes and their assigned geometry are summarised in Table 14, while the

electronic spectrum for complex 137a is illustrated in Figure 37.

Table 14: Electronic absorption bands with their assigned transition types and geometries for the cobalt complexes.

Complex Absorption (nm) Assignment Geometry

133 600sh,613,657,678 4A2 ~ 4T1(P) tetrahedral

134 579sh,608,628 4A2 ~ 4T1(P) ~ tetrahedral

136 576,621,638 4A2 ~ 4T 1(P) tetrah ed ral

137a 596sh, 604, 665, 677 4~ ~ 4T1(P) tetrahedral

137b 597sh, 611, 656, 676 4A2 ~ 4T1(P) tetrahedral

137c 580sh, 608, 629, 677sh 4.~ ~ 4T1(P) ~ tetrahedral

138 597sh , 604, 666, 678 4A2 --> 4T1(P) tetrahJd ral

In all cases, absorption spectra were observed corresponding to the 4A2 ---t 4T1(P)

transition, which is characteristic of tetrahedral cobalt coordination. The "fine

structure" evident in Figure 37 is attributed to spin-orbit coupling effects and to

transitions involving doublet states. 230

Discussion 109

"----.-- --------. - -~ - - -----_.--

-- --.. -.---------- ---~

Figure 37: The electronic spectrum of complex 137a in DMF.

,I

Discussion 110

2.3.4.1.3 Computer modelling of the cobalt complexes

These studies were undertaken to explore the possible 3-D structures of the

dinuclear cobalt complexes because crystals suitable for single crystal X-ray

analysis could not be obtained. Experimental IR and UV-Visible spectroscopic data

has been taken into account in developing representative structures of the

complexes detailed in Figure 38. For some of these complexes, isomeric structures

of the product are possible and molecular mechanics calculations were used to

predict the most likely structure. Thus, the symmetrical complex 138c, is

considered more likely than the isomeric systems 138a or 138b.

137b

.1

138c

Figure 38: Cobalt complexes modelled.

The computer modelled structure of the 1,1 O-phenanthroline diamide cobalt

complex 138c is shown in [Figure 39(a)]. In this model, the distorted tetrahedral

geometry of the metal ions can again be clearly seen. For the 1,1 O-phenanthroline

diamide complexes coordination through 'the carbonyl oxygen of the amide

functionality is indicated (see section 2.3.4.1.2). Complex 137b was modelled on

· ,-

Discussion 111

this basis and the resulting structure illustrates the conformation expected to be

typical of the analogous diamide cobalt complexes 137a and 137c. In this model ,

the distorted tetrahedral geometry of the metal ions can be seen [Figure 39(b)] .

(a)

(b)

I

Figure 39: Computer-modelled structures of the 1,1 O-phenanthroline cobalt

complexes: (a) complex 138c and (b) complex 137b for the biphenyl

diamide cobalt complexes. Coloured atoms: carbon: light'blue;

nitrogen: dark blue; oxygen: red; chlorine: green; cobalt light-brown.

112

139

Discussion

-N

~ N N

Ha~ ~aH CaJ \ ( ~a~ HO OH

-N CI N­'\ I /

CO N----I ~N

Ha~ CI ~aH CoJ \ ( ~O~

HO OH

112

Figure 40: The dendrimer-based ligand 112 and the proposed structure of its cobalt(lI) complex 139. \

I .

It was also decided to model the dendrimer-based ligand 112 and the proposed

cobalt(lI) complex 139 (Figure 40). From the models shown in Figure 41, the 3-D

arrangement of the dendrimer moieties in the ligand and the compl~x can be

clearly seen. In the cobalt complex (Figure 41 b), the 1,1 O-phenanthroline spacer

is no longer planar, but slightly bent, while the octahedral geometry of the cobalt

cation is also clearly apparent.

'.

Discussion

(a)

(b)

I

Figure 41: Computer modelled structures of: (a) the dendrimer ligand 112; (b) the resulting cobalt(lI) complex 139.

113

Discussion 114

2.3.4.2 Nickel Complexes

Various oxidation states have been observed for nickel ranging from (-I) to (+IV),

but studies of the chemistry of nickel have concentrated on the (+11) oxidation

state. The metal ion in nickel(lI) complexes exhibits considerable variability in

molecular geometry, e.g. four coordinate (tetrahedral or square planar), five

coordinate (square pyramidal or trigonal bipyramidal) and six-coordinate

(octahedral)?31 As a consequence, the chemistry of nickel(lI) is quite complicated.

Tetrahedral complexes are intensely blue in colour and paramagnetic;232 square

planar complexes are diamagnetic and usually red, brown or yellow in colour.232

Thus, tetrahedral complexes may be distinguished from square planar complexes

by their colour and their magnetic properties. Octahedral complexes may also be

blue or purple in colour and paramagnetic,due to the dB ion having two unpaired

electrons. 232 The nickel(llI) oxidation state is rare and very few Ni(lll) complexes

are known; the (+IV) oxidation state, which is the highest for nickel, is also rare.

Nickel plays an important role in biological systems. The first such system to be

discovered was the enzyme, urease, isolated from jack beans233.234 and shown by

UV-Visible spectroscopy and EXAFS studies to contain two octa~edral nickel(lI)

ions in a nitrogen- and oxygen-donor environment. Urease is responsjble for

catalysing the hydrolysis of urea to ar.nmonia and carbonic acid, thus:-

Nickel has subsequently been found in other enzymes, viz., hydrogenases, CO

dehydrogenases and coenzyme F 430. 234 Hydrogenases which occur in many

bacteria, catalyse the oxidation of molecular hydrogen, while CO dehydrogenases

interconvert carbon monoxide and carbon dioxide. It has been found that the

-. - hydrogenases and CO dehydrogenases have nickel(rII) (low spin d 7) and Fe-S

clusters present in their stwctures.234 Coenzy;me F 430, which is a cofactor of methyl

coenzyme M reductase and which contains nickel(lI) in a square planar geometry,

- . -

Discussion 115

is functional in a series of reactions in bacteria that result in the generation of

methane gas.

2.3.4.2.1 Synthesis of nickel complexes

Nickel chloride hexahydrate (NiCI 2.6H20) was used to prepare the nickel

complexes (Schemes 24 and 25), following the procedure detailed in the

Experimental section (3.3.3).

From the microanalysis data summarised in Table 15, it can be seen that:- i) most

of the complexes examined are dinuclear; ii) most of the complexes have four

chloride ions, except the mononuclear complex 143 for which there are two; iii)

complexes 141 and 142, appears to have.. DMF present, possibly as part of an

octahedral coordination sphere; iv) deprotonation of the amide, amine, imidazole

or benzimidazole groups has not occurred during formation of the dinuclear

complexes.

Table 15: Microanalytical data for the nickel(lI) complexes followE(d, in parentheses, by the calculated values

J

Complex Complex

Stoichiometry

% Carbon % Hydrogen % Nitrogen

140 Ni2(55a)CI4·6Hp 40.5 (41.1 ) 4.2 (4.7) 7.0 (6.9)

141 Ni2( 55b )CI4(DMF)4' 9Hp 41.7 (42.6) 5.3 (5.9) 11.7 (113)

142 Ni2( 55e )CI4(DMF)4 3H2O 41.4 (41.8) 5.2 (5.7) 13.2 (135)

143 Ni(61a)CI21H2O 57.1 (56.5) 4.6 (5.0) 6.7 (6.3)

144a Ni2(69a)CI4.2H2O 42.6 (43.6) 4.0 (3.7) 11.3 (10.9)

144b Ni2(69b )CI4(DMFh .8H2O 41.3 (42.1) 4.1 (4.8) 13.4 (12.9)

144e Ni2(6gelfI4· 2Hp 37.6 (38.5) 4.2 (3.5) 14.5 (15.0)

145 Ni2(68a)CI4·3Hp 44.1 (44.2) 4.0 (4.5) 11.0 (11.0) ---------------------------------------------------------------------------------------------------------------------------

55a-c

R

a

b

c

61a

Discussion

NiCb.~~~ ./ Me7"MeoH

DMF

NiC".6H~

MeCN

142

~~

I: N~ ~ CI " CI

116

140

141

I

143

Scheme 24: Complexation reactions of the biphenyl ligands with nickel(II).

NH

S R

0 HN

( R

69a-c

68a

Discussion

a b c

R --.LiN, N~ N~ ~+tN-0~HN

NiCI2.6H:zO or • DMF

144a-c

or MeCN

145

0

CI

Scheme 25: Complexation reactions of the 1 ,1 O-phenanthroline ligands with nickel(ll).

117

I

Discussion 118

2.3.4.2.2 Spectroscopic analysis of the nickel(lI) complexes

The 1 H NMR spectra of the biphenyl and the 1,1 O-phenanthroline nickel(lI)

complexes were expected to permit paramagnetic (tetrahedral and octahedral) and

diamagnetic (square planar) nickel(ll) complexes to be distinguished. However, the

1 H NMR spectra of the 1,1 O-phenanthroline complexes typically reveal a

combination of sharp and broad peaks as illustrated by the spectra of complex

144c and the ligand 69c (Figures 42 and 43). In the 1H NMR spectra of the

biphenyl Ni(lI) complexes, the peaks in the aromatic region (06-8 ppm) were,

typically, better resolved than in the spectra of the 1,1 O-phenanthroline Ni(lI)

complexes. Nevertheless, the signals for the biphenyl complexes are still broader

than the corresponding free rigand signals (cf. Figures 44 and 45). These

observations precluded unambiguous assignment of the metal ion geometries.

IR studies were undertaken to determine whether coordination with nickel

occurred through the oxygen or the nitrogen of the amide group, since it is known

that nickel coordinates with both oxygen and nitrogen donors. Pertinent IR data are

summarised in Table 16.

Table 16: Summary of the IR frequencies (VM_C,) and the amide frequency shifts (.6.VNH and !1Vc=o) on formation of the nickel complexes.

Complex .6.VNH Icm-1 .6.vc=o/cm -1 V M-c,/cm

-1

140 15 -11 276, 325

141 11 -6 387

142 61 13 378

144a -33 2 253, 378

144b -78 23 283,384

144c -~

-61 0 310,373

The 1,1 O-phenanthroline amide complexes 144a and 144b, exhibit negative shifts

of the amide NH band an'd,-positive shifts fo(the amide carbonyl bands indicating

coordination through the amide nitrogen. In the case of complex 144c no shift

Discussion 119

I I

pp. "

Figure 42: The 400 MHz 1 H NMR spectrum of the 1,1 O-phenanthroline complex 144c in DMSO-ds"

o I{N~ HN~i

J

69c

S Jj~ J J

pp. 10 ~ I , t

3

Figure 43: The 400 MHz 1 H NMR spectrum of the 1,1 O-phenanthroline ligand 6ge in DMSO-ds"

Discussion

__ L] ppo 12 10

Figure 44: The 400 MHz 1H NMR spectrum of the biphenyl complex 140 in DMSO-ds"

55a HWI ~N ~ 0 f:I~ ~N 0 ~ N....vv

I ~ H

I

JJ/J 10 8 ~ . " ";' Ii""""""": PPII 12 .q . -

I"'"'' , • I I" • 0

Figure 45: The 400 I'y1ljz 1 H NMR spectrum of the biphenyl ligand 55a in DMSO-ds" <

120

Discussion 121

for the amide carbonyl is observed, but a negative shift of the amide NH band is

an indication of coordination through the amide nitrogen. For the biphenyl amide

complexes 140 and 141 coordination through the amide oxygen is indicated by

positive NH shifts and negative carbonyl shifts; in the case of complex 142,

coordination through the amide nitrogen is suggested by the positive carbonyl

shift, although /). VNH is positive. The presence of the amide NH band in all cases

shows that deprotonation of the amide nitrogen has not occurred, thus supporting

the microanalysis data for these complexes.

A benzimidazole secondary amine NH stretching band was observed in the IR

spectra of complexes 141 and 144b; NH stretching bands for the imidazole

secondary amine were also evident in the IR spectra of complexes 142 and 144c,

further supporting the microanalysis data which, in each case, indicated the

presence of four chloride anions.

In the case of complexes 141 and 142 the strong bands in the 370-390 cm-1 region

are attributed to chloride anions coordinated in a trans geometry in an octahedral

environment. For complex 143, the two relatively strong bands observed at 289

and 329 cm-1 are believed to be due to chloride ions coordinated within a

tetrahedral environment, and the X-ray crystal structure for complex 143 Wigure

46) clearly confirms the tetrahedral geometry. Very weak bands at 276 and 325

cm-1 in the far IR region were observed for 140 and have been tentatively aSSigned

to Ni-CI stretches. The low intensity of these bands suggests that the chloride

anions are not in a tetrahedral environment (in contrast to those observed for

complex 143) and that the geometry of complex 140 is, in fact, square planar. In

the case of complexes 144a-c, the strong bands in the region 370-390 cm-1 are

attributed to trans-coordinated chloride ions in an octahedral environment, while

others at 253,283 and 31 0 cm-\ are indicative of chloride ions coordinated in a

- tetrahedral environment.235 Similar bands for complex 145 (at 278 and 352 cm-1)

are also considered to indiCate trans-coordinated chloride ions in tetrahedral and

octahedral environments respectively.

Discussion 122

J

Figure 46: X-ray crystal structure of complex 143 showing the crystal/ographic numbering.

Discussion 123

Different UV-Vis absorption patterns characterise different geometries in nickel

complexes,231,236,237,238 and the absorption data, proposed transition types and

geometries for some of the nickel complexes synthesised in this study are

summarised in Table 17. The absorption bands for the complexes 144b and 144c

appear to be masked by solvent bands (DMF and DMSO); for complex 145 a

shoulder at 377 nm has been tentatively assigned as the main tetrahedral

absorption band. The proposed geometries apply to at least one, but not

neccessarily both, Ni2+ ions in each dinuclear complex.

Table 17: Electronic absorption bands with their assigned transition types and the proposed geometries for the nickel complexes.

Complex Absorption (n m) Assignment Geometry

140 425 "'io. -- Square planar

141 412 3A29

-. 3T29 Octahedral

590 3A29

--> 3T19

(F)

619 3A29

--> 3T19

(P)

142 410 3A29

-> 3T29 Octahedral

525 3A29

-, 3T,i F)

576 3A29

--> 3T,9

(P)

143 514.555 3T, --> 3T,(P) Tetrahedral

144a 363 3A29

--> 3T29 Octahedral \

582 3A29

--> 3T19

(F)

621 3A29

--> 3T,9

(P)

144b masked -- ---144c masked --- ---

145 377sh 3T, --> 3T,(P) Tetrahedral

I

Discussion 124

2.3.4.2.3 Computer modelling of the nickel complexes

In the absence of X-ray crystal structures, computer modelling was used to explore

the possible 3-D structures of the dinuclear nickel complexes. The microanalytical,

IR and UV-Visible spectroscopic data were again considered in developing

structures for these complexes. In some cases, isomers are possible and

molecular mechanics calculations have been used to predict the most likely

structure. In this section, the discussion will focus on the computer modelling of the

representative dinuclear nickel complexes shown in Figures 47 and 49.

141

142

CI

_ --Ni--- 1 ~ H DMF I DMQy /1 ~ ~ N NH o-~ CI \)o~. 0 ~ h

HN ______ N I ~ "-.1 /' ~-H ~ /Ni"",--

DMF 1 DMF CI

CI DMF // DMF-Ni

/I-N~ N~NH H ~

o I

Figure 47: The representative biphenyl nickel complexes that have been modelled.

In the case of the biphenyl diamide complexes, coordination is presumed to involve

amide oxygen in complex 141 (to give 8-membered chelate rings) and the amide ~~

nitrogen in complex 142 (to give 6-membered chelate rings). The .energy-

minimised models of these complexes are shown in Figure 48. The octahedral,

metal ion geometry and ltfe mutually perpendicular arrangement of the arene rings

is apparent in these complexes.

'.

Discussion

(a)

(b)

I

Figure 48: Energy-minimised models of (a) complex 141 and (b) complex 142. Coloured atoms: Carbon: light-blue; nitrogen: dark-blue; oxygen: red; chlorine: green; nickel: light-green.

125

Discussion

(a)

(b)

, ,-

I

126

Figure 49: Computer-modelled structures of isomeric possibilities for the nickel(lI) complex 144a.

For the 1,1 O-phenanthroline complex 144a two isomeric products are possible [(a)

and (b); Figure 49]. These two complexes were modelled and the potential energy

of each structure was calculated and found to be -89.6 kcalmol-1 for isomer (a) and

40.5 kcalmol-1 for isomer (b) , suggesting that isomer (a) is the favoured _structure.

In each case, one of the nickel(ll) ions is octahedral , the other tetrahedral.

Discussion 127

2.3.4.3 Platinum complexes

Platinum exhibits a very unusual pattern in transition metal chemistry. Although the

other transition metals display a range of oxidation states, platinum is one of the

few that display oxidation states differing by two electrons, viz., 0, II and IV.239

For Pt(IV), low-spin, kinetically inert d 6 octahedral complexes are obtained,240

while Pt(lI) affords low-spin, kinetically inert d 8 square planar, diamagnetic

complexes.239 Although very few complexes of Pt(O) are known, examples do exist

in which platinum is two, three and four coordinate; in this oxidation state, the

ground state can be considered to be d10 or d9s 1 and coordination with arsine,

phosphine and isocyanide ligands has been reported. 239 The less common

oxidation states (I) and (III) have assuf!1ed some importance due to their

intermediacy in substitution reactions. 239

In addition to its widespread use in catalysis, jewelry and electrical applications,

platinum has assumed particular importance in the form of its complex, cisplatin,

the cis isomer of [Pt(NH3h(Clhl Cisplatin is used as an anti-cancer drug for

treating several kinds of malignant tumours.239 It is highly biologically active, while , the trans isomer is ineffective. The discovery of cisplatin as a chemotherapeutic

agent has resulted in the rapid development of bio-platinum chemistry.241

2.3.4.3.1 Synthesis of the Platinum Complexes

The platinum complexes of the biphenyl (Scheme 26) and 1,1 O-phenanthroline

ligands (Scheme 27) were generally prepared by adding a solution of the ligand

in DMF or MeCN, to a solution of K2[PtCI4] in water. The microanalysis results

(Table 18) obtained for these platinum complexes show that they are dinuclear and

_. Jhat four chloride ions are present in all cases except complex 149 in which there

are two, suggesting that the amide group is deprotonated in this complex. The ~ .- ~

expected benzimidazole complexes 147 and 150 did not appear to form, being

Discussion

inconsistent with the structures depicted in Schemes 26 and 27.

55a-c

H N~R

o

-9'1 N (GH

~ N ............... / Pt

/ ........ Cl CI

K2[PtCI4]

* DMF H20

a

146

b c

I

Scheme 26: Complexation of the biphenyl ligands with platinum:

128

149

150

Discussion

K:zIPtCI41 p=q .... o 0

DMF I'ItI HN

Hp ) ( R R

69a-c -x: K'PtCl,j DMF 1 K,PteI,j

H2O DMF H2O

AA i 2'2CI f' f' -N ~- N-

o \. / 0

~ ___ ~It~~~ \

N-Pt-N r ~ I ~ ~ HN CI NH

151

a b c

HN:)-r­~N

129

FRs HN I'ItI G8a

d b 1 K,PteI.

MeCN H2O

FB=f2

'

2C' f ~ f ~ -N N-

\. / __ pt __

b CIJ _N-~t-N ~!J I 'I ~ ~ CI _

152

J

Scheme 27: Complexation of the 1,.1 O-phenanthroline ligands with platinum.

Discussion

Table 18: Microanalytical data for the platinum complexes followed, in parentheses, by the calculated values.

130

Complex Complex

stoichiometry

% Carbon % Hydrogen % Nitrogen

---------------------------------------------------------------------------------------------------------

146 Pt2(55a)CI4·3Hp 34.6 (34.4) 3.2 (3.4) 7.0 (7.5)

148 Pt2( 55e )CI4.2Hp 29.7 (28.9) 3.4 (2.8) 7.8 (8.4)

149 Pt2(69a)CI2·3Hp 34.3 (34.6) 2.9 (2.8) 8.5 (B.7)

151 Pt2(6ge )C14 25.0 (25.9) 2.4 (5.1 ) 9.B (10.1 )

152 Pt2(6Ba)CI4·3Hp 31.B (31.1 ) 2.B (3.4) 7.3 (7.B)

2.3.4.3.2 NMR studies of the Platinum Complexes

For both the biphenyl and 1,1 O-phenanthrolme series, more signals are observed

than expected in both the lH and 13C NMR spectra of the resulting Pt(lI)

complexes. It was initially thought that n-stacking242 could be occurring in these

complexes; consequently, lH NMR spectra of the complexes were recorded at

60°C, but no changes in the spectra were observed, suggesting the absence of n­

stacking effects. The lH NMR spectrum of complex 149 (Figure 50) illustrates the

multiplicity of signals seen in the spectra of the 1,1 O-phenanthrGlline platinum

complexes - a feature which is particularly prominent in the aromatic region. The

signal multiplicity may reflect either a. certain asymmetry in the structures of the

complexes or, possibly, the presence of isomeric systems in solution.

, I PO'

, ,0 01

7

Discussion

, ~

o

Figure 50: The 400 MHz 1H NMR spectrum of complex 149 in DMSO-d6 o

69a

, ppm

, 12

-U '---- '- l..J. \.L ,

10

Figure 51: The 400 MHz 1 H NMR spectrum of ligand 69a in DMSO-d6 o

J

131

Discussion 132

2.3.4.3.3 IR studies of the platinum complexes

Platinum can coordinate to both nitrogen and oxygen and the IR studies were

expected to reveal which donor atoms in the biphenyl and 1,1 O-phenanthroline

amide complexes were involved in complexation. The IR data for relevant

absorption bands are summarised in Table 19.

Table 19: Summary of the IR frequencies (VM-C,) and the amide frequency shifts (LWNH and t;;vc=ot for the biphenyl and 1,1 O-phenanthroline Pt(lI) complexes.

Complex ~VNH /cm-1 ~Vc=o/cm-1 I -1 VM-cr em

146 -69 4 315,342 (cis)

148 31 - 4 335 (trans)

149 18 8 329 (trans)

151 -39 4 321 (trans)

a Frequency shift on complexation

From the data in Table 19 it appears that complexation does not involve

deprotonation of the amide group since an amide NH band is observed in each

case. It can also be seen that coordination through the amide nitrogeh is preferred

because positive shifts are observed for the amide carbonyl band in eacK case.

These results also indicate, in most cases, that trans coordination of the chloride

anion is favoured over cis coordination; in the case of complex 146, however, the

presence of weak Pt-CI bands at 342 and 315 cm-1 suggest a cis-geometry.243

The imidazole NH band is also present in the IR spectra of the biphenyl complex

148 (3120 cm-1) and the 1,10-phenanthroline complex 151 (3135 cm-1

) - an

observation in agreement with the microanalysis data which indicate four chloride

anions in each complex. For the biphenyl complex 148 the polymeric structure in

Scheme 26 has been tentatively suggested due to the trans coordination of the

chloride anion in this complex.

Discussion 133

The 1, 10-phenanthroline complex 152 exhibits a strong NH band at 3480 cm-1,

suggesting that the secondary amine is not deprotonated upon complexation - a

conclusion which is consistent with the microanalysis data obtained for this

complex.

The geometry of the platinum complexes could not be elucidated by UV-Vis

spectroscopy due to masking of the absorption bands of interest by the solvent,

DMF.

J

Discussion 134

2.3.4.3.4 Computer modelling of the platinum complexes

Since none of the dinuclear platinum complexes afforded crystals suitable for

single crystal X-ray crystallography, computer modelling studies were undertaken

to explore their possible 3-D arrangements. It was also hoped that the results of

these modelling studies might explain the multiplicity of signals observed in the 1H

and 13C NMR spectra of the 1,1 O-phenanthroline platinum complexes. The

specimen complexes which were modelled are shown in Figure 52.

146

149

J

Figure 52: Proposed structures of the biphenyl and 1,1 O-phenanthroline platinum complexes investigated by computer modelling.

The energy-minimised structures for the two specimen systems are shown in

Figure 53. The structure of complex 149 reveals a trans arrangement of the

chlorine ligands in a squ£lre planar environment, while the biphenyl complex 146

exhibits a cis-arrangement of the chlorine ligands -- observations which are

consistent with deductions based on the IR absorption data. The apparent

symmetry in the model of the 1, 10-phenanthroline complex 149, however, is not

consistent with the multiplicity of peaks observed in the aromatic region of the

:.

Discussion 135

(a)

(b)

Figure 53: Computer modelled structures:- (a) complex 149; (b) complex 146. (Colours: Carbon: light-blue; nitrogen: dark-blue; oxygen: red ; chlorine: green; platinum: lilac). J

spectra of the 1,1 D-phenanthroline complexes. It is, of course, possible that the

use of a universal force constant for platinum resulted in structures which reflect

a degree of symmetry not present in the actual complexes.

Discussion 136

2.4 ELECTROCHEMICAL STUDIES: CYCLIC VOL TAMMETRY

Electrochemical techniques can be used to study the redox properties of

organometallic complexes. The synthetic dinuclear complexes were subjected to

a particular electrochemical technique called cyclic voltammetry, to establish the

oxidation state(s) and explore the electronic effects of the ligands. This technique

also reveals the reversibility or irreversibility of a redox process when the redox

potential of a redox couple is measured.

IZOjIA

1" 0.:5 - 0.:5

VOLTS

.1

Figure 54: A cyclic voltammogram for a reversible redox process.

In cyclic voltammetry, the potential is scanned linearly from an initial value Ej to a

second value Ej and then back to Ej, and the current monitored. The peaking in

current occurs at the potential Epc where the species of interest is reduced or

oxidised. For reversible systems, the re-oxidation or re-reduction of the species

in the vicinity of the electrode gives a current peak in the reverse scan, the

- . potential corresponding to this peak is Epa in Figure 54. For irreversible systems,

no return peak is obsefV§d. The separation, between the cathodic and anodic peak

potentials, .1E (Epa -Epc), is equal to 59/n, where n is the number of electrons

Discussion 137

required in the redox process for a reversible system. Disproportionation reactions

cause irregularities in the shape of the curves in a cyclic voltammogram. 244 The

redox potentials are, however, dependent upon various factors such as:

- solvent effects

- the electronic nature of the ligands and their substituents

- the steric demands of the ligand

- the flexibility of the ligand and the size of the chelate rings in the complex245

In the following discussion, an attempt has been made to explain the results of the

cyclic voltammetric studies obtained for the copper, cobalt and nickel complexes

of selected ligands.

2.4.1 Copper Complexes

The analysis of the dinuclear copper complexes by cyclic voltahlmetry is important

as the results may indicate the biomimetic potential of the synthetic models of the

enzyme, tyrosinase. If reversibility is observed in the cyclic voltammogram of a

synthetic model, then biomimetic activity may be expected; if no~ reversibility is

observed, then the catalytic activity is likely to be poor or absent. J

Since it is known that copper(l) is unstable and easily oxidised to copper(ll) it is

important to establish whether the synthetic models can maintain a copper(l)

oxidation state and whether they permit the oxidation to copper(II). Dinuclear

copper(lI) complexes may be reduced in a single two-electron reduction, as

observed for a triketonate complex,246 or in a stepwise manner with a Cu(l)Cu(lI)

species being formed as an intermediate. Generally, it has been found that

synthetic dinuclear copp~er complexes are reduced in a stepwise manner by two -

-. - single-electron steps. "Type III" copper sites in proteins are, however, reduced in

a single two-electron redt:lction at redox potentials that are higher than expected

for their proposed coordination environment. 246 To quantify the redox processes,

Discussion 138

the standard redox potentials are measured and the separation, ~E, between the

consecutive reduction potentials, is determined, i.e.

where ~E is a measure of ~om' the comproportionation constant, which is the

equilibrium constant for the reaction:

CU"CU" + CUICUI -t CUICU"

and ~om is related to ~E by the equation:247

For the dinuclear complexes examined here, the copper ions could, in principle,

both be copper(I), both copper(II), or a combination of copper(l) and copper(II).

The biphenyl and 1~ 10-phenanthroline complexes

From the cyclic votammograms, no reduction to CUD was evident during the 1

analysis of the biphenyl and 1,1 O-phenanthroline copper complexes, except for ,I

complex 116. Peaks that could not be assigned to either the ligand or the copper

oxidation states, may be ascribed to ligand oxidation induced by the presence of

copper. Such ligand oxidations have complicated the assignment of peaks in the

cyclic voltammograms of the copper complexes and have also precluded the

assignment of oxidation states for most of the copper complexes. Moreover, the

polymeric nature of most of the copper complexes has resulted in complicated

cyclic voltammograms. Nevertheless, an attempt has been made to assign some,

but not all, of the peaks observed. The measured potentials of the copper

complexes are summarised in Table 20.

Discussion 139

Table 20: The measured potentials for the biphenyl and 1,1 O-phenanthroline copper complexes.

Complex Potentials measured in cyclic voltammogram (V) vs Ag/AgCI

115 Oxidation side: 0.32 (Ea' Cu2+/Cu3+)

Reduction side: -0.20 (Ea, ligand), -0.46 (Ea, EV2 = -0.51) and -0.55 (Ee, E112 = -0.51, CUIiCUIl

+ 2e- ---> (CUICUI).

116 Reduction side: -0.29 (Ea), -0.55 (Ee) and -0.80 (Ee).

118a Oxidation side: 0.25 (Ea)

Reduction side: -0.46, -0.52 (Ee' ligand), -0.65 (Ee, ligand), -0.97 (Ea, E112 = 1.1), -1.13 (Ee,

E112 = 1.1, CUIiCUIl + 2e- ---> (CUICUI).

118b Oxidation side: 0.85 (Ea)' 0.48 (Ea)'

Reduction side: -0.41 (Ee, ligand), -0.61 (Ee), -1.17 (Ee), -1.67 (Ee' ligand).

118c Oxidation side: 0.86 (Ea) ~

Reduction side: -0.50 (Ee, CUIICUII + 2e----> CUICUI ), -1.57 (Ee, ligand).

119 Oxidation side: 0.75 (Ea, ligand), 0.34 (Ee, CUIiCUIl ---> CUIllCUIll), 0.14 (Ee, ligand).

Reduction side: -0.71 (Ea, E112 = -0.77) and -0.82 (Ee, E112 = -0.77, CUIiCUIl + 2e- ---> CUICUI).

For complex 115, the reduction CUIiCUIl - CUICUI was assigned to the slightly

reversible couple at -0.46 and -0.55 V, and the oxidation peak at 0.32 V to a ~

CU2+/CU3+ oxidation. For complex 116, the sharp peak at -0.29 V has been ,I

assigned to copper deposition at. the electrode [i.e. the formation245,248 of

copper(O)]; the sharp peak is typical for absorbed species. The peak observed at

-0.80 V was assigned to a CUIiCUIl - CUICUI reduction and the one at -0.55 V to a

CUICUI - CuoCuo reduction, which corresponds to a two-step, two-electron

reduction process.

For complex 118a, the peaks at -0.52 and -0.65 V have been ascribed to the

ligand, while the the coufJle at -0.97 and -1.13 V (E1/2 = -1.1 V) has been assigned

- to a CUIiCUIl - CuICu' reduction.138 For complex 118b, it is not certain whether the

peak at -1.67 V is due)o CUll - CUi re9uction or whether it is due to ligand

reduction. It is also not possible to tell whether the peaks at -0.61 V and -1.17 V

are due to Cu" - CUi reduction, but they were not observed for the ligand. The

peaks at 0.85 and 0.48 V are, however, attributed to ligand oxidation. For complex

Discussion 140

118c, the peak at -0.50 V has been tentatively assigned to to a CUll -. CUi

reduction. The peak at -1.57 V cannot be assigned with certainty as a copper

peak, and it may well be due to ligand reduction. For this complex, only one Cu(lI)

oxidation peak (0.86 V) is observed, which suggests that both copper atoms are

in a irreversible Cu(lI) oxidation state.

A peak at 0.34 V in the cyclic voltammogram of complex 119 may be due to a

CUIiCUIl -. CulilCuliioxidation, and the couple at -0.71 and -0.82 V to a single two­

electron Cu"cd -. CJCu reduction; the LiE of -0.11 V suggests a quasi-reversible

one-electron system. The peaks observed at 0.75 and 0.14 V may be attributed to

ligand oxidation, by comparison with the ligand cyclic voltammogram.

The Schiff base complexes and the macro cyclic complex

The formation of copper(O) and its adsorption on the electrode is characterised by

a sharp peak in the cyclic voltammogram. 248 The cyclic voltammogram of each of

these copper complexes, but not their respective ligands, displayed this

characteristically sharp peak. The measured potentials for these copper complexes

are shown in Table 21.

Table 21: The measured potentials of the Schiff base and the J

macrocyclic copper complexes.

Complex Potentials measured in cyclic voltammogram (V) vs Ag/AgCI w

120 Reduction side: -0.26 (Ea) , -0.98 (Ee) and -1.40 (Ee), -1.75 (Ee)'

121 Oxidation side: 0.76 (Ea)' *

Reduction side: -0.22 (Ea), -0.59, -0.75 (Ee).

123 Oxidation side: 0.75 (E.). *

Reduction side: -0.13 (Ea), -1.00(Ee)' -1.75 (Ee)' w

124 Reduction side: -0.28 (Ee), -0.61 (Ee)' ..

*Peaks assigned to CUICul• CUD and copper deposition

Several peaks were obseFVed for complex 120. The strong, sharp peak at -0.26 V

has been assigned to copper deposition at the electrode. The peak at -0.98 V is

Discussion 141

attributed to a Cullcdl -t CUICUI reduction and the peak at -1.40 V to a CUICUI

-t CuoCuo reduction, thus representing a two-step, two-electron reduction

process138 to CUD. The peak at -1.75 V is assigned to ligand reduction by

comparison with the ligand cyclic voltammogram. For complex 121, the cyclic

voltammogram also exhibits a strong, sharp peak at -0.22 V, and this is attributed

to copper deposition. The peaks at -0.59 and -0.75 V have been assigned to a

two-step, single-electron reduction process and the peak at 0.76 V to ligand

oxidation. Complex 123 also exhibits similar peaks to those of complexes 120 and

121. The strong, sharp peak at -0.13 V has been assigned to copper deposition,

and the peak at -1.75 V to ligand reduction. Here, however, the two-step, single­

electron reduction process was absent since only one peak was observed at -1.00

V, and it is suggested that this peak may be due to a single, two-electron reduction

process. The peak· at 0.75 V has been ascribed to ligand oxidation. The

macrocyclic complex 124 exhibits a strong peak at -0.28 V, which is attributed to

copper deposition, but the peak at -0.61 V could not be assigned with certainty and

it seems from a comparison with the ligand cyclic voltammogram, that this latter

peak is due to ligand reduction.

A general mechanism reported in the literature for the reduction of copper(lI) to , copper(O) has been proposed138 to follow the sequence:-

I

2~r 2e-CUIlCUIl ---t Cu'Cul ---t 2Cuo

In this sequence, the two copper(lI) ions are reduced to copper(l) by a potential of

-0.98 V, followed by a subsequent reduction to copper(O) by a potential of -1.40 V.

In turn, a potential of -0.26 V results in the re-oxidation of copper(O).

Discussion 142

2.4.2 Cobalt Complexes

In the case of the dinuclear cobalt complexes, cyclic voltammetry was used to

establish whether the metal ions were cobalt(lI) or (III) or a mixture of both. The

measured potentials for the cobalt complexes are shown in Table 22. From the

results it is apparent that cobalt(lI) is oxidised to cobalt(lIl) on the cyclic

voltammetric time scale. The observation of a strong peak at ca. 1 V, which was

not observed for the ligand alone, indicates that the oxidation involves the cobalt

ion. Since Co(lIl) cannot be oxidised any further, the results suggest that Co(lI) is

present and is oxidised to Co(lll).

When comparing the oxidation potentials of the biphenyl diamide complexes (133

and 134) with those of the 1,1 O-phenanthroline diamide complexes (137a, 137b

and 137c), it is apparent that the latter complexes have higher oxidation potentials.

It was also observed that the 1,10-phenanthroline diamine complex 138 has a

lower oxidation potential than the diamide complexes, and it is presumed that this

is due to electron-withdrawal by the carbonyl oxygen of the amide group. Thus,

complex 138 is more easily oxidised than the diamide complexes, complex 137a

proving to be the most difficult.

I

The higher oxidation potentials of. the 1,1 O-phenanthroline-based complexes

(compared to the biphenyl analogues) may be attributed to the presence of the

additional, electronegative nitrogen atoms of the 1,1 O-phenanthroline nucleus,

which increase the electron-withdrawing capacity of the ligands, thus increasing

the oxidation potential of the resulting complexes. The rigidity of the 1,10-

phenanthroline system and the smaller, less flexible 5-membered chelate rings

which are formed upon complexation (in contrast to the flexibility of the biphenyl

system and its formation of larger, more flexible 6-membered chelate rings) may

-. - also contribute to the higher oxidation potentials observed for the 1,10-

phenanthroline complexes. Representative examples of the cyclic voltammograms

obtained for the biphenyl and 1,1 O-phenanthroline cobalt complexes are illustrated

Discussion

in Figures 55 and 56.

Table 22: Measured potentials for the biphenyl and 1,1 O-phenanthroline cobalt complexes.

Complex Potentials measured in cyclic voltammogram (V) vs Ag+/AgCI

133 Oxidation side: 0.97 ( CO"CO" --t COII1COII1 )

134 Oxidation side: 0.99 (COIICOII --t Co'IICOIII ).

137a Oxidation side: 1.08 (COIICO" --t COII1COII1 ).

137b Oxidation side: 1.06 (CO"CO"--> Co'IIC011I ).

137c Oxidation side: 1.02 (COIICOII --t C01IIC011I ).

138 Oxidation side: 0.86 (COIICOII --t COIIICO"1).

143

Discussion 144

+30+I~~--~~~~~~~~--~~~~

J l O~ __ -~~ l i

-3°l r

1 ~ ~oJ ~

I ! 1 r

~Oi ~ I I l r

-120 +1 ---r----'I:---'--'-I -----'-----',:--'---'---1 -'---'1--""--'---1 -.-+-1 +1.4 +1.0 +0.6 +0.2 -0.2 -0.6 -1.0 -1.4

Potential ,V

Figure 55: Cyclic voltammogram of biphenyl cobalt complex 134.

+30 I I l ~

1

o~ I I- .I

I I 1 r <{ 1

:::J -30~ ~ ..... J 1 c: L .0)

I 1

L.. I L.. :::J ~ol r ()

~ I I l-

I I -gal r J ~ I

-120 I I , I , , I , , I

+1.4 +1.0 +0.6 +0.2 -0.2 -0.6 -1.0 -1.4

Potential,V

Figure 56: Cyclic voltammogram of 1,1 O-phenanthroline cobalt complex 137c.

Discussion 145

2.4.3 Nickel complexes

Cyclic voltammetry was also used to establish the oxidation state(s) of the metal

ions in the dinuclear nickel complexes. Although the nickel(lI) oxidation state is

the most stable and most prevalent, the possibility of the Ni(III) and Ni(IV) oxidation

states occurring within these complexes had to be considered. The cyclic

voltammetry data for the complexes is summarised in Table 23.

From these results it can be seen that nickel(lI) was oxidised to nickel(lIl) on the

cyclic voltammetric time scale, since only one oxidation peak was observed in

each case. 249 In certain complexes (141, 144a, 144b and 145), ligand reduction

is evidenced by the peaks between -1.13 and -1.25 V, the oxidation peak being

clearly visible in the cyclic voltammogram of complex 142 (Figure 57); the cyclic

voltammogram for the ligand SSc is illustrated in Figure 58.

The 1,1 O-phenanthroline-cobalt complexes exhibit higher oxidation potentials than

their biphenyl counterparts, this trend was not observed for the corresponding

nickel complexes. It is also apparent that, of the biphenyl systems, complex 142

(Figure 57) has the highest oxidation potential and 141 the lowest.

Of the 1,1 O-phenanthroline nickel .systems, complex 144b has the highest

oxidation potential and 144a the lowest. Although the ligand G8a has an oxidation

potential at 0.86 V (see Figure 60) it was expected that the complex 145 would

have a lower oxidation potential and, consequently, the peak at 0.85 V (Figure 59)

has been tentatively assigned to the oxidation of Ni2+.

Discussion 146

Table 23: Cyclic voltammetric data for the nickel complexes.

Complex Potentials measured in cyclic

voltammogram (V) vs Ag+/AgCI

140 oxidation side: 0.83 (Ni2+)

141 oxidation side: 0.80 (Ni2+/Ni3+)

reduction side: -1.25 (ligand reduction)

142 oxidation side: 0.84 (Ni2+/Ni3+)

144a oxidation side: 0.79 (Ni2+/Ni3+)

reduction side: -1.13 (ligand reduction)

144b oxidation side: 0.95 (Ni2+/Ni3+)

reduction side: -1.17 (ligand reduction)

144c oxidation side: 0.82 (Ni2+)

145 oxidation side: 0.85 (Ni2+/Ni3+)

reduction side: -1.13 (ligand reduction)

Discussion 147

+40 i-I -----L--L-------L------"----'-----"---LI -'-I ---1-1 --LI ----'-, --'--, ~I ----'---, --"---+1

l ~~ __ --------~L -40 l

1 ( I f ~ r

~ -120~ I

~ 1 ~ ~ ~oo~ l

l ~ ~OO~ ~

I I l r

-360 I 1 1 I I I'! 1 I I 1 I I I +1.6 +1.2 +0.8 +0.4 0 -0.4 -0.8 -1.2 -1.6

Potential,V

Figure 57: A cyclic voltammogram of biphenyl nickel complex 142.

+15 \ I

I I 1 ~

I

+5~ ~ J

I I 1 I <{

-5~ ~ ~ ..... J l c (]) I I l- I I-~ -151 r ()

~ ~ I I

-25l r

~ ~ -351 I

I I I I I I I

+1.4 +1.:.0 +0.6 +0.2 -0.2 -0.6 -1.0 -1.4

Potential, V - . -

Figure 58: A cyclic voltammogram of ligand 55e.

Discussion 148

+100 i I I I I I I I I

I --:dt l O~ ---I / / !

i

-< 1 / I ::l -100~ / ~ ..... J l c: Q)

I I L. L. I ::l

-2ool I ()

~ ~

-300~ I I I

~ ~ I

-400 I I

! I I I I I I I I

+1.6 +1.2 +0.8 +0.4 0 -0.4 -0.8 -1.2 -1.6

Pote!!tial, V

Figure 59: A cyclic voltammogram of the 1,1 O-phenanthroline nickel complex 145.

+10 I I

I -1 r I

-10~ L I I I

-< 1 I ::l -30~ ~ ..... J l c: Q) I I L. L. I I ::l -SOl I I ()

~ L I N

I I -7°1 r

~ ~ -90 I I

I I I I I I I

+1.4 +1.0 +0.6 +0.2 -0.2 -0.6 -1.0 -1.4

Potential,V

Figure 60: A cyclic voliammogram of the 1 ,1 O-phenanthroline ligand 6Sa that has an oxidation potential at 0.86 V.

Discussion 149

Conclusion

The cyclic voltammetry studies of the copper complexes reveal that the 1,10-

phenanthroline ligands are unstable once they are coordinated to copper, ligand

oxidation being apparent in the cyclic voltammograms of the complexes. In the

case of the biphenyl analogues, complex 115 exhibits a Cu(ll) oxidation state,

while in complex 116 the metal is reduced to Cu(O), resulting in copper deposition

at one of the electrodes. In the Schiff base and macrocyclic complexes the copper

is also reduced to Cu(O) with concomitant electrodeposition of Cu(O). As a result

of the complexity of the cyclic voltammograms and the electrodeposition, it has

been difficult to assess either the redox reversibility or the potential of the copper

complexes as biomimetic catalysts.

In the cobalt complexes Co(lI) is oxidised to Co(llI) on the cyclic voltammetric time

scale, indicating that cobalt is in the Co(lI) oxidation state in the biphenyl- and

1,10-phenanthroline-based cobalt complexes. Similarly, the nickel complexes

exhibit predominant oxidation of the metal to Ni(lIl) confirming the oxidation state

of nickel in the biphenyl- and 1,1 O-phenanthroline-based complexes as Ni(II).

J

- .

Discussion

2.5 EVALUATION OF THE CATALYTIC ACTIVITY OF THE BIOMIMETIC

COPPER COMPLEXES

150

In order to determine whether the copper complexes were suitable models of

tyrosinase, their ability to oxidise a phenol and/or catechol was examined. For a

model to be successful it should display either phenolase or catecholase activity

using molecular oxygen. For this study, the substrate di-t-butylphenol (DTBP) 153

was used to evaluate the phenolase activity, and di-t-butylcatechol (DTBC) 154 the

catecholase activity of these complexes. 15, 250 These substrates were used because

they are activated for electron-donation by their alkyl substituents,208 and their

oxidation products are somewhat resistant to polymerisation. DTBP is oxidised to

DTBC which, in turn, is oxidised to 3,5-di-t-butyl-o-quinone (OTBQ) 155. The 0-

quinone has a characteristic absorption ban(j at 400 nm in the UV-visible spectrum

and can therefore be detected spectroscopically (e = 1830 M-1cm-1). DTBP can

also be oxidised to the coupled product 156.

\ (y0~ ~

153

OH

154

o

155 156

The reactions for evaluating biomimetic activity were conducted in DMF or in

CH2CI2. DMF was chosen because it is a highly polar solvent and was.expected

to dissolve, at least, some of the complex. Reactions were also attempted in

CH2CI2 because of its capacity to dissolve dioxygen, which must be present in non­

rate limiting concentrations. Tyrosinase can catalyse the oxidation of a variety of

substrates in an aqueous medium, but the range of substrates is decreased when

an organic medium is used.251 In this study, 1H NMR spectroscopy was ljsed to

detect the product(s) of oxidation reactions catalysed by selected biphenyl and

1,1 O-phenanthroline copper complexes.

Discussion 151

The 1H NMR data for the expected oxidation products DTBC 154 and DTBQ 155

and the coupled product 156 are detailed in Table 24, while the results of the

biomimetic evaluation of the copper complexes are summarised in Table 25. At the

conclusion of each reaction period, the reaction mixture was concentrated to

dryness in vacuo and the residue analysed by 1 H NMR spectroscopy to establish

the substrate: product ratios.

Table 24: 1H NMR data for DTBC 154, DTBQ 155 and the coupled product 156.

Compound OJppm

DTBC 154 1.27, 5, 9H, But

1.41, S, 9H, But

6.76, d, 1 H, H-4

~.89, d, 1 H, H-6

DTBQ 155 1.22, 5, 9H, But

1.26, S, 9H, But

6.21, d, 1 H, H-4

6.90, d, 1H, H-6 -

Coupled product 156 1.31, s, 18H, But

1.41,5, 18H, Bd

7.10, d, 2H, H-4 and H-4'

7.38, d, 2H, H-6 and H-6'

I

. Table 25: Results of the biomimetic evaluation of the copper complexes in

DMF.

Reaction Phenolase % Catecholase % Entry Complex Solvent time/h activity Conversion activity Conversion

product product

* 1 115 DMF 24h none none o-quinone 6.3

2 116 DMF 120 h coupled product 8.8 o-quinone 13.3

* 3 118a DMF 24h coupled product 14.3 o-quinone 7.3

4 118b DMF ~ 120 h coupled product 13.0 o-quinone 38.0

5 118c DMF 120 h none none" o-quinone 54.3

6 119 DMF 120 h none none none none

* Stirred for 24 h without ~tl.J

Discussion 152

The catalytic oxidation reactions were first attempted in CH2CI2, in the absence of

Et3N, but these were unsuccessful. The reactions were then repeated in DMF, but

catalytic activity was only observed for certain complexes. The complexes which

still failed to display any catalytic activity were then reacted, in DMF, in the

presence of EtJi138 Under these conditions, three more complexes exhibited

biomimetic activity (Entries 2, 4 and 5; Table 25). From the tabulated results, it can

be seen that some complexes displayed both phenolase and catecholase activity

(Entries 2,3 and 4), while all the complexes, with the exception of complex 119,

exhibited catecholase activity.

..

Scheme 28: Mechanism for the oxidative phenolic coupling proposed by Kitajima.20

Signals for the coupled product 156, indicating phenolase activity, were thus

observed in reactions with complexes 116, 118a and 118b. The formation of the

coupled product 156 has been observed by Reglier138 and Karlin,253 and other

coupled biphenyl products from reactions of phenol$ and biomimetic -copper

complexes have also been widely reported.254-257 Kitajima has proposed a

mechanism for the phenolic coupling, involving catalytic oxidation (Scheme 28).20

156

I pplfl 7 5

I 7.'

Discussion

, 7.3 7.2

153

/1

H-4 l __ ~~~~~.~

1:1 [::1

7.' 7.0

Figure 61: Partial 1H NMR spectrum, in CDCI3 , of the residue following reaction of complex 118a and DTBP in DMF in the absence of ET3N, showing the formation of the coupled product 156, i.e. phenolase activity.

155 ±~o xUo

. ( ,I \11 I J

H~I H-4 .

II! II I :\ I !~~W t 1 I~~I \

___ ~0J \J~J ~ Iii" I

PPII 10 , 1

Figure 62: The 400 MHz 1 H NMR spectrum, in CDCI3 , of the residue following reaction oJ c;omplex 118c an<~ DTBC in DMF in the presence of Et3N, showing formation of the ortha-quinone 155, i.e. catecholase activity.

Discussion 154

Catecholase activity was confirmed by the 1H NMR signals for the o-quinone at ca.

06.2 and 06.9 ppm observed in the spectra of residues from the reactions with

complexes 115,116, 118a, 118b and 118c) .

It is possible that the lack of phenolase activity exhibited by some of the complexes

may be due to either the polymeric copper complexes being too flexible to bind

DTBP 153, or to coordination being limited to only one copper atom, thus making

a-hydroxylation unlikely. The fact that both phenolase and catecholase activity is

exhibited by some of the complexes suggests that, in these cases at least, the

copper ions are sufficiently close (after binding dioxygen) to allow for the binding

and bridging of DTBP 153 and DTBC 154. While the copper complexes appear to

be polymeric (see Section 2.3.1.1.), the formation of polymer-monomer equilibria

in solution would afford some monomer moiecules capable of acting as biomimetic

catalysts. On the other hand, the macromolecular characteristics of tyrosinase

may also have been simulated, to some extent, by the polymeric structure of the

biomimetic complexes.

I

Discussion 155

2.6 CONCLUSIONS

During the course of this research, several sets of ligands, designed to mimic the

active site in tyrosinase, have been successfully prepared. The diimine ligands

containing the biphenyl spacer have been shown to undergo intramolecular

cyclisation during reduction with sodium borohydride, affording azepine

derivatives; X-ray crystallographic analysis of cobalt and nickel complexes

facilitated identification of the azepine systems, and a mechanism has been

proposed to account for the observed cyclisation. In contrast, neither the 1,10-

phenanthroline nor the Schiff base analogues undergo intramolecular cyclisation.

However, Schiff base ligands, in which the imine functionalities are separated by

one carbon atom, undergo significant tautomerism, forming six-membered H­

bonded chelates.

Complexation of the biphenyl- and 1,1 O-phenanthroline-based amide ligands with

copper appears to afford polymeric copper complexes, analysis of which reveals

coordination through both the amide nitrogen and oxygen donor atoms, as well as

through the tertiary and secondary nitrogen donors of the imidazole and

benzimidazole moieties. In the case of the Schiff base copp$r complexes,

unidentate and bidentate coordination of the nitrate anion has complicated

elucidation of the structures.

Tetrahedral geometry appears to be dominant in the dinuclear cobalt complexes

of the biphenyl- and 1,1 O-phenanthroline-based amide I igands. Coordination with

nickel, results in the formation of complexes in which the metal exhibits tetrahedral,

octahedral or square planar geometries, while the biphenyl and 1,10-

phenanthroline platinum(ll) complexes appear to favour a trans arrangement of

the chlorine ligands in a square planar environment.

Unlike the polymeric copper complexes, the cobalt and nickel complexes afforded

uncomplicated cyclic voltammograms, confirming the formation of cobalt(ll) and

Discussion 156

nickel(l/) complexes with the biphenyl- and 1,1 O-phenanthroline-based ligands.

Despite the polymeric nature of the copper complexes, biomimetic activity for both

biphenyl- and 1,1 O-phenanthroline-based polymeric complexes was observed. The

presence of triethylamine appears to enhance the biomimetic activity, and it seems

that the copper atoms in the complexes are close enough for dioxygen bridging

to occur and that substrate binding is possible. Such evidence suggests the

existence of polymer-monomer equilibria in solution, with the monomeric dicopper

complexes capable of acting as biomimetic catalysts.

Future research in this area is expected to involve:-

i) complexation of the biphenyl- and 1,1 O-phenanthroline-based amide ligands

with copper in the presence of base; -

ii) complexation of the Schiff base and macrocyclic ligands with copper(l) to

explore the potential of the resulting complexes as biomimetic catalysts;

iii) an investigation of the potential of the biphenyl- and 1,1 O-phenanthroline­

based cobalt(lI) complexes to bind oxygen and thus catalyse the oxidation of

3,5-di-t-butylphenol and 3,5-di-t-butylcatechol; and

iv) exploration of the potential of the nickel(l/) complexes as catal~sts for the

hydrolysis of urea. J

157

3 EXPERIMENTAL

3.1 GENERAL

Infrared spectra were recorded on Perkin Elmer 2000 and Perkin Elmer 180

spectrophotometers; the mid-infrared spectra (4000-300 cm-1) were recorded using

potassium bromide discs and, nujol mulls and hexachlorobutadiene (HCBD). The

far-infrared (500-50 cm-1) spectra were obtained using nujol mulls. NMR spectra

were recorded on a Bruker AMX 400 spectrometer, and chemical shifts are

reported relative to the solvent peaks. Low resolution mass spectra were obtained

on a Hewlett-Packard 5988A mass spectrometer, and high resolution analyses on

a Kratos MS80RF double focussing magnetic sector instrument (Cape Technikon

Mass spectrometry unit); FAB mass spectra~were obtained on a VG Micromass 70-

70E spectrometer (Ion tech B11 N FAB-gun), using Xe as bombarding gas

(University of Potchefstroom). UV-Visible spectra were recorded on a Cary 1 E UV­

Visible spectrometer, and the resulting data for the cobalt and nickel complexes

are summarised in Tables 14 and 17 respectively. Microanalysis (combustion

analysis) was conducted at the University of Cape Town, and the data for the

copper, cobalt, nickel and platinum complexes are reported in Tables~8, 12, 15 and

18 respectively. X-ray crystallographic data were collected on a Siemens sMART

CCD diffractometer at the University of the Witwatersrand. Electrochemical data

were recorded with a Bio Analytical Systems (BAS) CV-50 voltammograph. Melting

points were obtained using a Kofler hot-stage microscope and are uncorrected.

Experimental 158

3.2 PROCEDURES FOR LIGAND SYNTHESES

3.2.1 Biphenyl ligands

Biphenyl-2,2'-dicarba/dehyde 48140

A stirred suspension of phenanthrene 47 (2.5 g, 14 mmol) in dry methanol (50 mL)

was cooled to -30°C in an ozonolysis vessel. 0 3 was gently bubbled through the

mixture until all of the substrate had dissolved. KI (8.5 g) and glacial AcOH (7.5

mL) were added to the mixture at O°C. The mixture was then allowed to stand at

room temperature for 1 h, after which, an aqueous solution of Na2S20 3 (10%) was

added. A stream of air was passed through the mixture for 2 h. Cold H20 (O°C; 50

mL) was added, and the solid product was filtered off and recrystallised from Et20-

hexane to give, as a pale yellow solid, biphenyl-2,2'-dicarbaldehyde 48 (2.74 g,

93%), mp 61-62°C [lit. 140 62-63°C]; vmruJKBr/cm-1) 1700 (CO); oH(400MHz; CDCI3)

7.35 (2H, d, ArH) 7.59 (2H, t, ArH), 7.64 (2H, t, ArH), 8.05 (2H, d, ArH) and 9.83

(2H, s, CHO).

Diphenic Acid 49141

Phenanthrene 47 (10 g, 56 mmol) was dissolved in glacial AcOH (113 mL) and the

solution warmed to 85°C on a water bath. An aqueous solution of H20 2 (BO%;39

mL, 0.45 mmol) was added dropwise during a period of 40 minutes after which the

mixture was heated for a further 3-4 h. The volume of the mixture was then

reduced to half by distillation under reduced pressure, and the diphenic acid,

which precipitated out, was filtered off. The filtrate was evaporated almost to

dryness under reduced pressure and then extracted through warming with an

aqueous solution of Na2C03 (10%; 42 mL). This extract was boiled with a little

decolourising carbon and then filtered. To the filtrate, dilute HCI was added until

a pH of 4.5 was reached (using narrow range indicator paper). The solution was

-. - further stirred with a small amount of activated charcoal and the tarry material was

filtered off. The clear solution was then acidified with dilute HCI at O°C. The crude

diphenic acid was collected by filtration, washed with H20, dried at 110°C and then

Experimental 159

recrystallised from glacial AcOH. The total yield of diphenic acid 49 was 9.1 g

(67%), mp 228-229°C [lit.141 230°C]; vmaiKBr/cm-1) 2500-3500 (OH stretch), 1700

(CO); oH(400 MHz; DMSO-d6) 7.18 (2H, d, ArH) 7.43 (2H, t, ArH), 7.53 (2H, t, ArH),

7.93 (2H, d, ArH) and 12.44 (2H, s, C02H).

2-(2-Aminoethy/)benzimidazo/e 52138

A stirred mixture of 1 ,2-diaminobenzene 57 (8.33 g, 77 mmol) and .B-alanine 58

(10.2 g, 115 mmol in HCI (6M; 90 mL) was boiled under reflux for 48 h, and then

allowed to stand for 48 h at room temperature. The completion of the reaction was

confirmed by TLC. The solvent was removed under reduced pressure at 90°C, and

the residual solid was dissolved in H20 (50 mL). EtOH (100 mL) was then added,

and the dihydrochloride salt crystallised out as blue needles on cooling.

Recrystallisation from 95% EtOH gave~c the dihydrochloride (4.0 g, 32%). A

methanolic solution of NaOMe (0.1 M; 171 mL, 17.1 mmol) was then added to the

salt (2.0 g), and the mixture stirred for 1.5 h. The solvent was removed under

reduced pressure before adding CHCI3 (50 mL) to dissolve the amine. The

solution was filtered, and a pale pink product was obtained after removing the

CHCI3 under reduced pressure. Recrystallisation from THF gave 2-(2-amino­

ethyl)benzimidazole 52 (1.1 g, 80%), mp 134 -135°C (lit. 138 134 -136°C);

vrnax (KBr/cm-1) 3350 (NH) and 1540 (C=N); oH(400 MHz; 0 20) 2.91 (2H, ,t, CH2),

2.99 (2H, t, CH2), 7.21 (2H, m, ArH),}.49 (2H, m, ArH).

Histamine 53

Histamine dihydrochloride (2.0 g, 11 mmol) was added to a methanolic solution of

NaOMe (0.1 M; 220 mL, 22 mmol) and the resulting mixture stirred for 1.5 h at room

temperature. The solvent was then removed under reduced pressure, and CHCI3

was added to the residue. The flask was warmed to aid the dissolution of the

liberated amine; the solution was filtered, and evaporation of the solvent gave, as

_. - a light yellow oil, histamine 53 (1.20 g, 98%); vrnax(thin film/cm-1) 3481 and 3250

(NH); oH(400MHz; COCht2.72 (2H, t, CH2), 2.99 (2H, t, CH2), 4.5 (2H, br s, NH2)'

6.79 (1H, s, ArH) and 7.53 (1H, s, ArH).

Experimental 160

2, 2'-Bis[4-(2-pyridyl}-2-azabut-1-eny/]bipheny/ 54a 103

2-(2-Aminoethyl)pyridine 51 (1.1 mL, 9.0 mmol) was added to a solution of 1,1 '­

biphenyl-2,2'-dicarbaldehyde 48 (1.00 g, 4.76 mmol) in CHCI3 (50 mL) in a 2-

necked flask fitted with a condenser and a modified Dean-Stark apparatus. The

mixture was boiled under reflux for 16 h, and completion of the reaction was

confirmed by IR spectroscopy. The CHCI3 was removed under reduced pressure

to afford, as a pure brown oil, 2,2'-bis[4-(2-pyridyl)-2-azabut-1-enyl]biphenyl 54a

[(1.88 g, 97%); vmax (NaCllcm-1) 1650 (C=N); oH(400 MHz; CDCI3) 3.03 (4H, t,

NCH2CH2), 3.76 (4H, m, NCH2C~), 7.00 (4H, t, ArH), 7.06 (2H, d, ArH), 7.34 (4H,

m, ArH), 7.47 (2H, t, ArH), 7.85 (2H, s, CH=N), 8.04 (2H, d, ArH), 8.40 (2H, d,

ArH)], which was used without further purification.

2, 2'-Bis{[2-(2-pyridy/)ethy/amino]carbonyJ}bipheny/ 55a

~ ~N 0

o ~~ ~ H

Diphenic acid 49 (1.00 g, 4.1 mmol) was dissolved in dry DMF (10 rnL) in a round­

bottomed flask fitted with a reflux condenser and drying tube. The soluti'on was

warmed to 40°C, and COl (2.13 g, 13.1 mmol) was added with stirring. The mixture

was stirred for 5 min at 40°C, after which time, gas evolution ceased. After cooling

to room temperature, 2-(2-aminoethyl)pyridine 51 (1.1 mL, 9.0 mmol) was added.

The resulting solution was stirred for 1 h at room temperature, and the reaction

was quenched with H20 (7 mL). Volatiles were removed under reduced pressure,

and aqueous 1 M Na2C03 (50 mL) was added to the residual oil. The mixture was

extracted with EtOAc (2 x 80 mL), and the combined extracts were washed with

H20 (80 mL) and brine (8D mL), and dried (MgS04). The solvent was evaporated

-. - and the residue chromatographed [flash chromatography on silica gel; elution with

CHCI3-hexane-MeOH -(3:3:1)] to afford, as a brown oil, 2,2'-bis{[2-(2-

pyridy/)ethyJamino]carbonyl}bipheny/55a (1.00 g, 54%) (Found: M+, 450.2044.

Experimental 161

C2sH2S02N4 requires, M 450.2054); vmaithin film/cm-1) 3321 (amide NH), 1634

(CO); oH(400 MHz; CDCI3) 2.68 (4H, quintet, NHCH2CH2), 3.42-3.67 (4H, m,

NHCH2CH2), 7.02-7.12 (6H, m, ArH), 7.25-7.35 (4H, m, ArH), 7.48 (2H, m, ArH),

7.52-7.58 (2H, m, ArH), 7.74 (2H, t, NH), 8.45 (2H, d, ArH); oC<100 MHz; CDCI3)

37.1 (NHCH2CH2), 39.1 (NHCH2CH2), 121.7 ,123.6,127.4,128.0, 129.6,129.7,

136.6, 136.8, 139.4, 149.4 and 159.6 (ArC) and 170.1 (CO).

2,2'-Bis{[2-(2-benzimidazo/y/)ethy/amino]carbonyl}bipheny/55b

Following the procedure used for synthesising 2,2'-bis{[2-(2-pyridyl)ethylamino]­

carbonyl}biphenyl 55a, a solution of diphenic acid 49 (1.00 g, 4.1 mmol), 2-(2-

aminoethyl)benzimidazole 52 (1.46 g, 9.0 mmol) and COl (2.13 g, 13.1 mmol) in

dry DMF (10 mL) was stirred for 64 h at room temperature. The reaction was

quenched with H20 (7 mL), and the solvent~as removed under reduced pressure.

Addition of aqueous Na2C03 (1 M; 50 mL) to the residual oil precipitated the crude

product, which was recrystallised from DMF-H20 to afford, as a pale pink powder,

2,2'-bis{[2-(2-benzimidazo/y/)ethylamino]carbonyl}bipheny/55b (1.21 g, 56%) mp

> 250°C (Found: M+, 528.2261. C32H2SNs02 requires, M 528.2272; vmax(KBr/cm-1)

3172 (amide NH), 1656 (CO); oH(400 MHz; DMSO-d6) 2.62 (4H, s, NHCH2CH2),

3.43 (4H, s, NHCH2CH2), 6.99 (2H, d, ArH), 7.05-7.16 (4H, m, ArHh7.28-7.37 (4H,

m, ArH), 7.38-7.42 (4H, m, ArH), 7.49 (2H, d, ArH), 8.64 (2H, m, amide NH~, 12.14

(2H, s, NH); oc(100 MHz; DMSO-d6)t 28.2 (NHC~C~), 37.2, (NHC~C~), 110.7,

118.0,120.7,121.4,126.9,127.2,128.9,134.2,136.2, 138.5,143.2 and 152.1

(ArC) and 169.0 (CO).

2,2'-Bis{[2-( 4-imidazolyl)ethylaminojcarbonyljbiphenyl 55e

The procedure described for the synthesis of 2,2'-bis{[2-(2-pyridyl)ethylamino}­

carbonyl}biphenyl 55a was followed, using histamine 53 (1.00 g, 9.0 mmol), COl

(2.13 g, 13.1 mmol) anadiphenic acid 49 (1.0 g, 4.1 mmol). After stirring for 94 h -

-. - at room temperature in dry DMF (10 mL), the reaction was quenched with H20 (7

mL) and the volatiles -were removed under reduced pressure. The addition of

t The coincidence of some 13C signals is presumed.

Experimental 162

aqueous 1 M Na2C03 (50 mL) to the residual oil precipitated the crude product,

which was recrystallised from DMF-H20 to afford, as an off-white powder, 2,2'­

bis{[2-(4-imidazo/y/)ethy/amino]carbonyl}bipheny/55c (1.44 g, 82%), mp 232-234°C

(from DMF-H20) (Found: M+, 428.1955. C24H2402Ns requires, M 428.1961);

vmax(KBr/cm-1) 3187 (amide NH), 1640 (CO); oH(400 MHz; MeOH-d4) 2.51 (4H, s,

NHCH2CH2), 3.34 (4H, s, NHCH2CH2) , 6.73 (2H, s, ArH), 7.08 (2H, m, ArH), 7.40

(4H, m, ArH), 7.48 (2H, m, ArH), 7.53 (2H, s, ArH); od100 MHz; MeOH-d4) 27.5

(NHCH2CH2),40.5(NHCH2CH2), 117.9, 128.4, 128.9, 130.6, 130.7, 135.7, 136.1,

137.4 and 140.4 (ArC) and 172.5 (CO).

2,2'-Bis{[2-(2-pyridy/)ethy/amino]methy/}bipheny/ 56a

Attempted method 1.

H ~N ~N ~

~

1.9 N H

2,2'-Bis[4-(2-pyridyl)-2-azabut-1-enyl]biphenyl 54a(1.00 g, 2.39 mmol) was added

to methanol (10 mL) in a 2-necked flask fitted with a reflux conpenser. To the

heated mixture, NaBH4 (0.18 g, 4.8 mmol) was added in portions while ,ptirring.

Once the addition was complete, the,reaction mixture was boiled under reflux for

0.5 h and then quenched with ice. The product was extracted with CHCI3 (3 x 40

mL) and the combined extracts were dried over anhydrous MgS04. The CHCI3 was

removed under reduced pressure. A sample of the crude product was purified by

preparative thin layer chromatography [on silica; elution with MeOH-CHCI3-hexane

(1 :4:1)] to afford 1-[2-(2-pyridy/)ethyf]dibenz[c,e]perhydroazepine 61a as a brown

oil; oH(400 MHz; CDCI3) 2.97-3.01 (2H, m, NCH2CH2 ), 3.11-3.14 (2H, m,

NCH2CH2 ), 3.48 (4H, s, 2--x CH2 ), 7.10-7.13 (1 H, m, ArH), 7.23 (1 H, d, ArH), 7.33-

-. - 7.36 (4H, m, ArH), 7.40-7.44 (2H, m, ArH), 7.49 (2H, -d, ArH), 7.60 (1 H, td, ArH),

8.54 (1H, d, ArH); oc(1QO.MHz; CDCI3) 37.~ (NCH2CH2), 55.43 (NCH2CH2), 55.51

(ArCH2N), 121.2,123.2,127.5,127.6,128.0,129.7,134.7,136.3, 141.1, 149.3 and

Experimental 163

160.3 (ArC).

Attempted method 2.

NaBH4 (0.18 g, 4.8 mmol) was added in portions to a stirred solution of 2,2'-bis[4-

(2-pyridyl}-2-azabut-1-enyl]biphenyl 54a (0.46 g, 1.1 mmol) in EtOH (20 mL),

effervescence occurring during the addition of the NaBH4. The solution was stirred

for 67 h, at room temperature, before evaporating the EtOH under reduced

pressure to yield a yellow powder. This residue was dried under high vacuum to

remove remaining EtOH. CHCI3 was added to the residue to dissolve the product,

and the mixture was then filtered to remove unreacted NaBH4. The CHCI3 was

removed under reduced pressure to afford, as a pale brown oil, 1-[2-(2-

pyridyl)ethyl]dibenz[c,e]perhydro-azepine 61a (0.46 g, 100%).

2,2'-Bis{[2-(2-benzimidazolyl)ethylaminojinethy/jbiphenyI 56b

Attempted method 1.

A solution of biphenyl-2,2'-dicarbaldehyde 48 (0.49 g, 2.3 mmol) and 2-(2-amino­

ethyl)benzimidazole 52 (0.74 g, 4.6 mmol) in CHCI3 (100 mL) was boiled under

reflux for 59 h. The completion of the reaction was confirmed by TLC. The solvent

was removed under reduced pressure, and the residue recrystalised from MeCN

to afford the diimine 54b as a yellow powder (0.9 g, 80%). NaBH4 (0.27 g, 7.1

mmol) was then added to a solution of the diimine 54b in MeOH (10 mL), and the

mixture was boiled under reflux for 30-40 min. Ice was then added to the reaction

mixture to quench the reaction, precipitating, as a pale yellow powder, 1-[2-(2-

benzimidazolyl)ethyl)dibenz[c,ejperhydroazepine 61b (0.49 g, 65%) (Found: MH+,

340.1814. C23H21N3 requires MH, 340.1814), mp 184-186 °C; vmax(KBr/cm-1) 3174

(NH); 5H (400 MHz; MeOH-d4) 3.10 (2H, m, NHCH2CH2), 3.23 (2H, t, NHCH2CH2),

3.50 (4H, s, CH2NHC~C~), 7.17-7.21 (2H, m, ArH), 7.34-7.41 (4H, m, ArH), 7.44-

-. - 7.48 (2H, m, ArH), 7.49-7.52 (4H, m, ArH); 5d100 MHz; MeOH-d4)t 28.2

(NHCH2CH2), 54.4 (NHGH2CH2), 56.0 (CH2NHCH2CH2), 115.5, 123.4, 128.8,

129.1,129.8,131.2,134.7,139.4,142.5 and 154.7 (ArC).

Experimental 164

Attempted method 2.

Following the procedure described for the attempted preparation of 2,2'-bis{[2-(4-

imidazolyl)ethylamino]methyl}biphenyI55e (method 2), a stirred mixture of 2,2'-bis­

{[2-(2-benzimidazolyl)ethylamino]carbonyl}biphenyl 55b (0.50 g, 0.95 mmol) and

LiAIH4 (0.72 g, 19 mmol) was boiled under reflux in dry THF (50 mL) for 55 h under

N2. 1H NMR and IR spectroscopy of the material obtained after work-up indicated

that the desired product had not been obtained.

Attempted method 3.

A mixture of 2,2'-bis{[2-(2-benzimidazolyl)ethylamino]carbonyl}biphenyl 55b (1.0

g, 1.9 mmol) and Raney nickel (0.33 g) in absolute EtOH (60 mL) was stirred under

H2 for 48 h at room temperature. The reaction mixture was filtered through a celite

pad to remove the Raney nickel, which was quenched with HCI. The filtrate was

evaporated to dryness under reduced pressure. The residue was vacuum dried,

and analysis by 1 H NMR and IR spectroscopy indicated that this reaction was also

unsuccessful.

Attempted method 4.

A solution of 2,2' -bis{N-[2-(2-benzimidazolyl)ethyl]-N-benzamidol1lethyl}biphenyl

S3 (0.80 g, 1.1 mmol) in HCI (6N; 15 mL) was boiled under reflux for 14 h. The

reaction mixture was washed with Et20, basified with NaOH (10N) and then

extracted with CHCI3. The CHCI3 extracts were combined and then dried with

anhydrous MgS04. The CHCI3 was removed under reduced pressure to yield an

off-white powder, 1H NMR spectroscopy of which, showed that the reaction had

been unsuccessful.

2, 2'-Bis-{[2-( 4-imidazolyl)ethylaminojmethyl}biphenyl 5Se

-. - Attempted method 1.

Biphenyl-2,2'-dicarbaldebyde 48 (0.72 g, ~.4 mmol) was added to a solution of

histamine 53 (0.75 g, 6.8 mmol) in CHCI3 (100 mL), and the resulting solution

Experimental 165

boiled under reflux for 6 h, using a modified Dean-Stark apparatus to remove H20.

The solvent was removed under reduced pressure to yield a glassy yellow solid

shown, by 1H NMR spectroscopy, to be the impure diimine 54e (1.34 g, 99%)

[vmax(KBr/cm-1) 1650 (C=N)}. The crude product (1.00 g, 1.2 mmol) was then

dissoved in methanol (10 mL), and NaBH4 was added in portions. The resulting

mixture was boiled under reflux for 40 min. before adding ice to quench the

reaction. The precipitated product was washed with Et20 to afford, as a white

powder, 1-[2-(4-imidazo/y/)ethyljdibenz[c,ejperhydroazepine 61e (0.32 g, 33%), mp

149-151°C (Found: MH+, 290.165729. C1Ji19N3 requires, MH 290.1657;

vmax(KBr/cm-1) 3413 (NH); OH(400 MHz; MeOH-d4) 2.89 (2H, m, NCH2CH2), 2.95

(2H, m, NCH2CH2), 3.46 (4H, s, 2 x CH2), 6.83-6.87 (1 H, m, ArH), 7.37-7.44 (4H,

m, ArH), 7.45-7.58 (4H, m, ArH),7.59-7.63 (1H, m, ArH); Od100 MHz; MeOH-d4)t

56.08 (2x CH2), 56.35 (NCH2CH2), 128.77-129.0,129.6,131.1,135.2,136.0 and

142.5 (ArC).

Attempted method 2.

2,2'-Bis{[2-(4-imidazolyl)ethylamino]carbonyl}biphenyl 55e (0.80 g, 1.8 mmol) was

added in portions to a suspension of LiAIH4 (0.35 g, 9.2 mmol) in dry THF (40 mL),

and the stirred mixture was boiled under reflux for 8 h under N2 . Aqueous NaOH

(4M, 25 mL) was added to the reaction mixture, which was then stirred oljernight

to ensure quenching of the excess L.:iAIH4. The precipitated solid was filtered off

and washed with THF.1f The filtrate and the washings were combined and

evaporated to dryness under reduced pressure. 1H NMR and IR analysis of the

vacuum-dried residue indicated that the reaction was unsuccessful.

N-[(2-Benzimidazo/y/)ethy/]benzamide 62

A mixture of benzoic acid (1.00 g, 8.19 mmol), COl (2.16 g, 13.3 mmol) and 2-(2-

aminoethyl)benzimidazole (1.32 g, 8.19 mmol) in DMF (20 mL) was stirred for 2.5

_. - d following the procedure used for the preparation of 2,2'-bis{[2-(2-

pyridyl)ethylamino]carbonyl}biphenyl 55a. The pale pink powder, which

Experimental 166

precipitated out, was filtered off and washed with aqueous Na2C03 (1 M; 50 mL) to

give N-[(2-benzimidazo/y/)ethy/]benzamide 62 (1.58 g, 73%), mp >250°C (from

DMF-H20) (Found: MH+, 266.1294. C1sH1SN30 requires, MH 266.1293);

vmaiKBr/cm-1) , 3305 (NH), 3176 (amide NH),1638 (CO); oH(400 MHz; DMSO-d6)

3.09 (2H, t, NHCH2CH2), 3.73 (2H, m, NHCH2C~) ),7.12 (2H, m, ArH), 7.49 (5H,

m, ArH), 7.84 (2H, d, ArH), 8.63 (1H, m, amide NH), 12.50 (1H, s, NH); od100

MHz; DMSO-d6)t 28.7 (NHCH2CH2) , 37.9 (NHCH2CH2 ), 121.0, 127.0, 128.1,

131.0, 134.4 and 152.7 (ArC) and 166.2 (CO).

2, 2'-Bis-{N-[2-(2-benzimidazo/yl)ethyl]benzamidomethyl}biphenyl 63

To a stirred suspension of NaH (50% dispersion in oil, pre-washed with dry DMF

under N2; 0.19 g, 3.2 mmol) in dry DMF (10 mL), was added N-[(2-

benzimidazolyl)ethyl]benzamide 62 (0.78"'9,2.9 mmol), and the resulting mixture

was boiled under reflux for 1.5 h under N2. 2,2'-Bis-(bromomethyl)biphenyl (0.50

g, 1.5 mmol) was then added and stirring was continued at room temperature

under N2 for 2.5 d. The reaction mixture was heated for 2 h before working up. The

DMF was removed under reduced pressure before adding H20 to the residue. The

precipitated solid was filtered off, washed with H20 and recrystallised from DMF­

H20 to give, as an off-white solid, 2,2'-bis-{N-[2-(2-benzimidazclyl)ethyl]benz­

amidomethyl}biphenyl63 (1.01 g, 97%), mp >250 °C (Found: MH+, 709.329087.

C4sH40Ns02 requires, MH 709.329099); vmax(KBr/cm-1) 1638 (CO); OH(400 MHz;

DMSO-d6) 2.89 (4H, s, CH2C~NC~), 3.64 (4H, m, C~CH2NC~), 5.10-5.24 (4H,

m, CH2CH2NCH2), 6.67 (2H, d, ArH), 7.12 (4H, t, ArH), 7.06-7.17 (4H, m, ArH),

7.20-7.27 (4H, m, ArH), 7.27-7.34 (2H, m, ArH), 7.34-7.43 (6H, m, ArH), 7.45-7.50

(2H, m, ArH), 7.61 (2H, d, ArH) , 7.71 (4H, d, ArH) , 8.59 (2H, s, NH); oc(100 MHz;

DMSO-d6) 26.7 (CH2C~NC~), 37.3 (C~C~NC~), 44.6 (C~C~NC~), 110.0,

118.5,121.3,121.8,126.2,126.9,127.6,128.0, 128.3,129.7,130.9,134.17,134.2,

135.1,138.0,142.2 and 103.1 (ArC) and 166.2 (CO).

Experimental 167

3.2.2 1,1 O-Phenanthroline-based ligands

1, 1 O-Phenanthroline-2, 9-dicarbaldehyde 65145

A mixture of neocuproine 64 (3.0 g, 14 mmol) and Se02 (7.5 g) in dioxan (200 mL)

was boiled under reflux for 2 h, and then filtered through celite while hot. The

filtrate was cooled on ice, and the crude dialdehyde (2.5 g, 73%) separated from

the cold filtrate as a yellow powder. Recrystallisation from THF gave, as yellow

crystals, 1,1 0-phenanthroline-2,9-dicarbaldehyde 65, mp 231-232°C(Iit. 145 231-

232°C); vmax(KBr/cm-1) 1720 (CO); ~ (DMSO-q,) 8.28 (2H, s,Ar-H), 8.31 (2H, d, Ar­

H), 8.79 (2H, d, Ar-H), 10.36 (CHO).

1, 10-Phenanthroline-2,9-dicarboxylic acid 66145

A solution of 1,1 0-phenanthroline-2,9-dicarbaldehyde 65 (0.5 g, 2.1 mmol) and

80% HN03 (10mL) was boiled under reflux for 3 h. After cooling, the solution was

poured onto ice and the precipitated solid recrystallised from MeOH to give, as a

yellow solid, 1, 10-phenanthroline-2,9-dicarboxylic acid 66 (0.40-g, 65%), mp 238°C

(lit. 145 238°C); vmaiKBr/cm-1) 1724 (CO), 2800-3700 (COOH).

2,9-Bis{[2-(2-pyridy/)ethyljaminomethyl}-1, 10-phenanthroline 6Sa ,

To a suspension of 1, 10-phenanthroline-2,9-dicarbaldehyde 65 (1.12 g, 4.7'mmol)

in CHC13 (50 mL), was added 2-(2-aminoethyl)pyridine 51 (1.0 mL, 8.2 mmol), and

the resulting mixture boiled under reflux for 2 h. The completion of the reaction was

confirmed by IR spectroscopy. The CHCI3 was removed under reduced pressure

to yield the crude diimine 67a as a dark-brown oil (2.11 g, 95%), which became

semi-solid upon standing at room temperature. 1 H NMR and IR analysis confirmed

that the reaction had been successful. The semi-solid diimine 67a (1.0 9, 2.3

mmol) was dissolved in dry MeOH (10 mL), and NaBH4 (0.17 g, 4.51 mmol) was

added to the solution in portions. The resulting mixture was then reflLJxed for

-. -2.5 h, after which the reaction was quenched with ice. The product was extracted

with CHCI3 (4 x 40 mL) and the combined extracts were dried over anhydrous

MgS04. The CHCI3 was removed under reduced pressure to yield, as a brown oil,

- .

Experimental 168

which became a semi-solid upon standing at room temperature, 2,9-bis{[2-(2-

pyridy/)ethyljaminomethyl}-1,10-phenanthroline 68a (0.71 g, 71%) (Found: M+,

448.2371. C28H2SNe requires M, 448.2375; vmax (thin film/cm-1) 3300 (NH) and 1590

(C=N); oH(400 MHz; CDCI3) 3.08 (4H, t, CH2NHCH2CH2), 3.17 (4H, t,

CH2NHCH2CH~, 4.32 (4H, s, CH2NHCH2CH2), 7.06 (2H, t, ArH), 7.19 (2H, t,

ArH), 7.54 (2H, td, ArH), 7.72 (2H, d, ArH), 8.14 (2H, d, ArH), 8.49 (2H, t,

ArH); oC<100 MHz; CDCI3) 38.5 (CH2CH2NHCH2), 49.3 (CH2CH2NHCH2), 56.0

(CH2CH2NHCH2), 121.2, 122.0, 123.3, 125.8, 127.7, 136.3, 136.5, 145.2, 149.3,

160.3 and 160.8 (ArC).

2, 9-Bis-{[2-(2-benzimidazo/y/)ethyljaminomethyl}-1, 10-phenanthro/ine 68b

N~N H '~h

HN~_'

A solution of 2-(2-aminoethyl)benzimidazole 52 (0.74 g, 4.7 mmol) in MeOH (10

mL) was added dropwise to a ·hot solution of 1,10-phenanthroline-2,9-

dicarbaldehyde 65 (0.54 g, 2.3 mmol) in MeOH (30 mL). The resulting solution was

boiled under reflux for 1.5 h, completion of the reaction being confirmed by IR

spectroscopy. After removing the solvent under reduced pressure, the crude

diimine 67b was obtained as a glassy orange-brown solid. NaBH4 (0.29 g, 7.6

mmol) was then added in portions to a solution of the crude diimine 67b (1.0 g, 1.9

mmol) in MeOH (20 mL). After boiling the mixture under reflux for 45 min., the

reaction was quenched-with ice precipitating, as an orange-yellow powder, 2,9-

- bis{[2-(2-benzimidazo/y/)ethy/aminojmethyl}-1, 10-phenanthroline 68b, (0.40 g,

40%), mp >250°C (Found: MH+, 527.26~2. C32H30NS requires MH, 527.2672);

oH(400 MHz; DMSO-d6) 3.0-3.12 (8H, m, CH2NHCH2CH2), 4.16 (4H, s,

Experimental 169

CH2NHC~C~), 6.9-7.3 (4H, m, ArH), 7.30-7.60 (4H, br s, ArH), 7.87 (2H, d, ArH),

7.91 (2H, s, ArH), 8.41 (2H, d, ArH), 12.31 (2H, br s, NH); 5d100 MHz; OMSO-d6)t

29.3 (CH2NHCH2CH2), 47.2 (CH2NHCH2CH2), 55.0 (CH2NHCH2CH2), 120.9,

121.6, 125.7, 127.3, 136.4,144.4,153.8 and 160.8 (ArC).

Attempted Method

A stirred mixture of 2,9-bis{[2-(2-benzimidazolyl)ethylamino]carbonyl}-1, 1 0-

phenanthroline 69b (1.0 g, 1.7 mmol) and LiAIH4 (0.66 g, 17 mmol) in dry THF (40

mL) was boiled under reflux for 8 h under N2, and worked-up following the

procedure described for the attempted preparation of 2,2'-bis{[2-( 4-imidazolyl)­

ethylaminojcarbonyl}bipheny/55e (Method 2). 1H NMR and IR analysis of the

isolated material revealed that the reaction had been unsuccessful.

2, 9-Bis{[2-( 4-imidazo/y/)ethy/jaminomethyl}-1, 10-phenanthroline 68c

Attempted Method

To a suspension of LiAIH4 (0.36 g, 9.6 mmol) in THF (20 mL), 2,9-bis{[2-(4-

imidazolyl)ethylamino]carbonyl}-1, 1 O-phenanthroline 6ge (0.80 g, 1,9 mmol) was

added in portions. The resulting mixture was boiled under reflur for 9 h, and

worked-up following the procedure described for the attempted preparation,of 2,2'­

bis{[2-(2-imidazolyl)ethylamino]carbonyl}biphenyl 55e (Method 2). IR and 1 H NMR

spectra of the isolated material indicated that the reaction had not been

successful.

2, 9-Bis{[2-(2-pyridy/)ethy/aminojcarbony/}-1, 10-phenanthro/ine 69a

A solution of 1, 10-phenanthroline-2,9-dicarboxylic acid 66 (0.48 g, 2.05 mmol) in

dry OMF (10 mL) was heated to 40-S0°C. COl (1.07 g, 13.1 mmol) was added in

portions and the resulting-mixture stirred until effervescence ceased. 2-(2-Amino-

-. - ethyl)pyridine 51 (0.S5 mL, 4.S mmol) was then added at room temperature, and

the reaction mixture was stirred for 4.S days.<The reaction was quenched with H20

(7 mL), and the solvents were removed under reduced pressure at 90°C. Aqueous

Experimental 170

Na2C03 (1 M, 50 mL) was added to the residue, and the resulting precipitate was

filtered off to give, as a light-brown powder, 2,9-bis{[2-(2-pyridyl)ethy/amino]­

carbony/}-1,10-phenanthroline 69a (0.78 g, 73%), mp 55-57°C (from OMF-H20)

(Found: M+, 476.1960. C2sH24N602 requires M, 476.1961); vmax(KBr/cm-1) 3256

(NH) and 1651 (CO); oH(400 MHz; OMSO-d6) 3.18 (4H, t, NHCH2CH2),3.75-3.85

(4H, m, NHCH2CH2), 7.20 (2H, dd, ArH), 7.35 (2H, d, ArH), 7.69 (2H, t, ArH), 8.18

(2H, s, ArH), 8.44 (2H, d, ArH), 8.49 (2H, d, ArH), 8.73 (2H, d, ArH), 9.56 (2H, br

s, NH); OcC100 MHz; OMSO-ds) 36.9 (NHC~C~), 39.0 (NHCI1 CI1), 120.8, 121.6,

123.0,127.7,130.1,136.7,138.1,143.6,148.8,149.4 and 159.0 (ArC) and 163.7

(CO).

2, 9-Bis{[2-(2-benzimidazo/y/)ethy/amino]carbony/}-1, 10-phenanthroline 69b

A mixture of 2-(2-aminoethyl)benzimidazole 52 (1.46 g, 9.0 mmol), 1,10-

phenanthroline-2,9-dicarboxylic acid 66 (0.96 g, 4.1 mmol) and COl (2.13 g, 13.1

mmol) in dry OMF (20 mL) was stirred for 26 h, using the procedure described for

the synthesis of 2,9-bis-{[2-(2-pyridyl)ethylamino]carbonyl}-1,1 O-phenanthroline

69a. Work-up afforded as a light-brown solid, 2,9-bis{[2-(2-benzimidazo/y/)­

ethy/amino]carbony/}-1,10-phenanthro/ine 69b (1.53 g, 61 %), mp >250°C (from

OMF-H20) (Found: M+, 554.2179. ~~~ q requires M, 554.2179)i 'fuax (KBr/cm1 )

3145 and 3323 (NH) and 1630 (CO); oH(400 MHz, OMSO-d6) 3.29/(4H, t,

NHCH2CH2), 3.90 (4H, dd, NHCH2C~), 7.06-712 (4H,m, ArH), 7.41:'7.49 (4H, m,

ArH), 8.15 (2H, s, ArH), 8.45 (2H, d, ArH), 8.71 (2H, d, ArH), 9.71 (2H, br s, amide

NH); oc(100 MHz, OMSO-ds)t28.2 (NHC~C~), 37.8 (NHC~CH2)' 120.8, 121.2,

127.8, 130.1, 138.1, 143.5, 149.4 and 152.8 (ArC) and 163.6 (CO).

2, 9-Bis-{[2-( 4-imidazo/y/)ethy/amino]carbony/}-1, 10-phenanthroline 69c

Following the procedure described for the preparation of 2,9-bis{[2-(2-pyridyl)­

ethylamino]carbonyl}-1, 1Q-phenanthroline 69a, a mixture of histamine 53 (1.00 g,

_ . 9.0 mmol), 1, 10-phenanthroline-2,9-dicarboxylic acid-66 (0.96 g, 4.1 mmol) and

COl (2.13 g, 13.1 mmol) in dry OMF (15 mL) was stirred for 55 h. Work-up <

afforded, as a light-brown solid, 2,9-bis{[2-(4-imidazo/y/)ethyJamino]carbonyl}-1, 10-

Experimental 171

phenanthroline 69c (1.18 g, 63%), mp 122-124°C (from DMF-H20) (Found M+,

454.1851. C24~Ns~ requires M, 454.1866); vmax(KBr/cm-1) 3130 and 3297 (NH)

and 1649 (CO); oH(400 MHz, DMSO-d6) 2.93 (4H, t, NHCH2CH2), 3.68 (4H, dd,

NHCH2CH2), 6.91 (2H, s, ArH), 7.57 (2H, s, ArH), 8.17 (2H, s, ArH), 8.45 (2H, d,

ArH), 8.68 (2H, d, ArH), 9.55 (2H, t, amide NH); oc(100 MHz, DMSO-de)t 26.8

(NHCH2CH2), 39.3 (NHCH2CH2), 120.8, 127.7, 130.1, 134.7, 138.0, 143.6 and

149.5 (ArC) and 163.7 (CO).

J

Experimental 172

3.2.3 Schiff base ligands prepared from diketones and 2-(2-aminoethyl)­

pyridine

N, N' -Bis[2-(2-pyridyl)/ethyl]pentane-2, 4-diimine 76§

Acetylacetone 70 (0.41 mL, 4.0 mmol) was dissolved in CHCI3 (5 mL). To this

solution, 2-(2-aminoethyl)pyridine 51 (0.95 mL, 7.9 mmol) was added and the

resulting mixture was boiled under reflux for 2 h and then stirred at room

temperature for 48 h. The completion of the reaction was confirmed by TLC. The

solvent was removed under reduced pressure and the residue was purified by

chromatography [flash chromatography on silica; elution with MeOH-benzene­

EtOAc (1 :4:4)] to afford, as a brown oil, N,N'-bis[2-(2-pyridy/)ethyl]pentane-2,4-

diimine 76 (0.70 g, 57%) (Found: MH+, 309.2079. C19H24N4 requires MH,

309.2079); oH(400 MHZ; CDCI3) 1.82 (3H,cs, CH3), 1.95 (3H, s, CH3), 3.01 (4H, t,

NHCH2CH2), 3.65 (4H, q, NHCH2CH2), 4.89 (1 H, s, NHC=CH), 7.09-7.14 (2H, m,

ArH), 7.16 (2H, d, ArH), 7.58 (2H, t, ArH), 8.52 (2H, d, ArH), 10.85 (1 H, br s, 0 20

exchangeable, NH); od100 MHz; CDCI3) 18.6 (CH3), 28.7 (CH3), 38.8

(NHCH2CH2), 42.6 (NHCH2CH2), 95.2 (NHC=CH), 121.7, 123.6, 136.5, 149.5,

158.2, 162.8 and 194.7 (ArC, NHC=C and C=N).

3-methyJ-N, N' -bis[(2-(2-pyridyJ)ethyl]pentane-2, 4-diimine 77§

To a solution of 3-methyl-2,4-pentanedione 71 (0.41 mL, 3.5 mmol) in CHCI3 (5

mL), 2-(2-aminoethyl)pyridine 51 (0.87 mL, 7.0 mmol) was added, and the reaction

mixture was boiled under reflux for 2 h and then stirred at room temper~ture for 3

h. Completion of the reaction was confirmed by TLC. The CHCI3 was evaporated

off under reduced pressure. Purification of the residue by chromatography [(flash

chromatography on silica gel; elution with MeOH-benzene-EtOAc (1 :4:4)] yielded,

as a brown oil, 3-methyl-N,N'-bis(2-(2-pyridyl)ethyl]pentane-2,4-diimine 77 (0.48

g, 43%); (Found: MH+, 323.2235. C2oH26N4 requires MH, 323.2236); oH(400 MHz;

-. - CDCls) 1.78 (3H, s, CHs), 1.90 (3H, s, CH3), 2.08 (3H, s, CH3), 3.01 (4H, t,

NHCH2CH2), 3.66 (4H, q, .NHCH2CH2), 7.1 t (2H, m, ArH), 7.18 (2H,d, ArH), 7.58

§ In CDCls• the compound exists as a monoimino tautomer.

Experimental 173

(2H, td, ArH), 8.52 (2H, d, ArH), 12.00 (1H, br s, NH); oeC100 MHz; CDCI3) 14.7

(CH3)' 15.0 (CH3)' 28.4 (CH3), 39.0 (NHCH2CH2), 43.1 (NHCH2CH2), 98.1

(NHC=C), 121.6, 123.6, 136.5, 149.5, 158.5, 161.9 and 194.6 (ArC, NHC=C and

C=N).

3-Ethy/-N,N'-bis[2-(2-pyridy/)ethyljpentane-2,4-diimine 78§

A solution of 2-(2-aminoethyl)pyridine 51 (1.50 mL, 12.5 mmol) and 3-ethyl-2,4-

pentanedione 72 (0.84 mL, 6.2 mmol) in CHCI3 (5 mL) was boiled under reflux for

1 h. The reaction was shown to be complete by TLC. After removal of the solvent

under reduced pressure, the residue was chromatographed [flash chromatography

on silica; elution with MeOH-benzene-EtOAc (1 :4:4)J to give, as a brown oil, 3-

ethy/-N,N'-bis[2-(2-pyridy/)ethyljpentane-2,4-diimine 78 (0.73 g, 35%) (Found: MH+,

337.2392. C21H2SN4 requires MH, 337.2392); oH(400 MHz; CDCI3) 0.94 (3H, t,

CH 2CH3), 1.90 (3H, s, CH3), 2.10 (3H, s, CH3), 2.20 (2H, q, CH2CH3), 3.01 (4H,

NHCH2CH2), 3.65 (4H, q, NHCH2CH2), 7.10 (2H, t, ArH), 7.17 (2H, d, ArH), 7.57

(2H, t, ArH), 8.52 (2H, d, ArH), 12.14 (1H, br s, NH); oeC100 MHz; CDCI3) 14.3

(CH2CH3), 15.5 (CH2CH3) ), 22.0 (CH3)' 27.3 (CH3)' 38.9 (NHCH2CH2), 43.1

(NHCH2CH2), 105.5 (NHC=CH), 121.6, 123.6, 136.5, 149.47 ,149.49, 158.5, 162.2

and 194.6 (ArC, NHC=C and C=N).

3-benzy/-2,4-pentanedione 73

A stirred solution of acetyl acetone 70 (0.82 mL, 8.0 mmol) in methanolic NaOMe

(79.9 mL, 8.0 mmol) was boiled under reflux for 1.5 h. Benzyl bromide (0.95 mL,

8.0 mmol) was then added to the mixture, and stirring continued at room

temperature for 2.5 d. The solvent was evaporated off under reduced pressure and

the residue extracted with EtOAc. This mixture was then gently heated and filtered.

Evaporation of the solvent under reduced pressure gave, as a pale yellow oil, 3-

benzyl-2,4-pentanedione-?3 (0.92 g, 61%) [oc(100 MHZ; CDCI3} 29.7 (CH2), 30.0

-. - (CH3)' 72.1 (CH), 126.0, 128.21, 128.42 , 140.9 (ArC) and 207.8 (CO)], which was

used without further purjfi.cation.

Experimental 174

3-Benzy/-N, N' -bis[2-(2-pyridy/)ethy/]pentane-2, 4-diimine 79§

To a solution of crude 3-benzyl-2,4-pentanedione 73 (0.59 g, 3.1 mmol) in CHCI3

(5 mL) was added 2-(2-aminoethyl)pyridine 51 (0.75 mL, 6.3 mmol), and the

resulting mixture was boiled under reflux for 2.5 h. After removal of the solvent

under reduced pressure, the residue was chromatographed [flash chromatography

on silica; elution with MeOH-EtOAc (1:8)] to give, as a brown oil, 3-benzy/-N,N'­

bis[2-(2-pyridy/)ethyl]pentane-2,4-diimine 79 (0.35 g, 25%) (Found: MH+, 399.2548.

C26~N.t requires MH, 399.2549); ~(400 MHz;CDC6) 1.80 (3H, s, C~), 2.03 (3H,

s, CH3), 3.07 (4H, t, NHCH2CH2), 3.63 (2H, s, CH2Ph), 3.71 (4H, q, NHCH2CH2),

7.10 (2H, d, ArH), 7.11-7.18 (2H, m, ArH), 7.19-7.26 (4H, m, Ar-H), 7.61 (2H, td,

ArH), 8.54 (2H, d, ArH), 12.36 1 H, br s, NH); od100 MHz; CDCI3) 15.0 (CH3)'

27.9 (CH3), 34.6 (C~Ph), 39.0 (NHC~CH2)' 43.2 (NHCH2C~), 101.4 (CC~Ph),

121.7,123.8,125.7,127.5,128.4,136.5,141.7,149.6, 158.4, 163.7 and 195.4

(ArC, NHC=C and C=N).

1, 3-Diphenyl-N, N'-bis[2-(2-pyridyl)ethyl]propane-1, 3-diimine 80§

2-(2-Aminoethyl)pyridine 51 (0.85 mL, 7.1 mmol) was added to a solution of

dibenzoylmethane 74 (0.80 g, 3.6 mmol) in CHCI3 (10 mL), and the reaction

mixture was boiled under reflux for 6 d. Completion of the reaction was confirmed

by TLC. After removing the solvent under reduced pressure, the r~sidual material

was chromatographed (flash chromatography on silica gel; elution with BOAc) to

afford, as a yellow-green oil, 1,3-dipnenyl-N,N'-bis[2-(2-pyridy/)ethy/]propane-1,3-

diimine 80 (0.90 9, 58%) (Found: MH+, 433.2392. C29H28N4 requires MH,

433.2392); 0H(400 MHz; CDC~) 3.02 (4H, t, NHC~CH2)' 3.64 (4H, q, NHCH2C~),

5.69 (1H, s, NHC=CH), 7.09-7.17 (4H, m, ArH), 7.26-7.28 (4H, m, ArH), 7.33-7.43

(12H, m, ArH), 7.53-7.60 (2H, m, ArH), 7.30-7.87 (4H, m, ArH), 8.49 (2H, d, ArH),

11.41 (1 H, br s, NH); od100 MHz; CDCI3) 39.4 (NHCH2CH2), 44.4 (NHCH2CH2),

93.6 (NHC=CH),121.6, 123.6, 127.0, 127.6, 128.1, 128.5, 129.4, 130.6, 135.6,

136.4, 140.35, 149.5, 158.2, 166.7 and 188.4 (ArC, ~HC=C and C=N) ...

Experimental 175

1-Pheny/-N,N' -bis[2-(2-pyridy/)ethy/]butane-1, 3-diimine 81 §

To a solution of benzoyl acetone 75 (0.80 g, 4.9 mmol) in CHCI3 (10 mL) was added

2-(2-aminoethyl)pyridine 51 (1.2 mL, 9.9 mmol), and the reaction mixture was

boiled under reflux for 76 h. The reaction was monitored by TLC and completion

was confirmed. The CHCI3 was removed under reduced pressure, and the residue

was chromatographed [flash chromatography on silica; elution with MeOH-CHCI3-

hexane (1 :4: 1)] to afford, as a light-brown oil, 1-pheny/-N,N'-bis[2-(2-pyridy/)­

ethyl]butane-1,3-diimine 81 (0.77 g, 43%) (Found: MH+, 371.2235. C24H2SN4

requires MH, 371.2236); oH(400 MHz; CDCI3) 1.97 (3H, s, CH3), 3.09 (4H, t,

NHCH2CH2), 3.76 (4H, q, NHCH2C~), 5.60 (1H, s, NHC=CH), 7.14 (2H, m, ArH),

7.20 (2H, d, ArH), 7.33-7.43 (3H, m, ArH), 7.59 (2H, td, ArH), 7.82 (2H, m, ArH),

8.56 (2H, d, ArH), 11.49 (1 H, br s, NH); od100 MHz; CDCI3) 19.2 (CH3), 38.8

(NHCH2CH2), 42.9 (NHCH2CH2), 92.1 (f\!HC=CH), 121.7, 123.7, 126.8, 128.0,

130.3,136.5,140.4,149.6,158.2,164.7 and 187.7 (ArC, NHC=C and C=N).

N, N' -Bis[2-(2-pyridy/)ethyl]hexane-2, 5-diimine 84

2-(2-Aminoethyl)pyridine 51 (0.84.mL, 7.0 mmol) was added to a solution of

acetonylacetone 82 (0.41 mL, 3.5 mmol) in CHCI3 (5 mL), and the resulting mixture

was boiled under reflux for 2.5 d, completion of the reaction bein~ confirmed by

TLC. The solvent was removed under reduced pressure, and the residual rvaterial

was chromatographed [flash chromatography on silica gel; elution with MeOH­

EtOAc(1 :8)] to give, as a brown oil, N,N'-bis[2-(2-pyridy/)ethyl]hexane-2,5-diimine

84 (0.68 g, 60%) (Found: MH+, 323.2236. C2oH2SN4 MH, requires 323.2236);

vmax(NaCllcm-1) 1652 (C=N); ~(400 MHz; 2.13 (6H, s, CI1), 3.06 (4H, t, CIiC~ N),

4.13 (4H, t, CH2CH;N), 5.74 (4H, s, N=CCH2), 6.92 (2H, d, ArH), 7.12-7.18 (2H, m,

ArH), 7.55 (2H, td, ArH), 8.57 (2H, d, ArH); od100 MHz; CDCI3) 12.3 (CH3), 39.6

(CH2C~C=N), 43.4 (CH2CH2C=N), 105.2 (NCH2CH2), 121.7, 123.6, 127.4, 136.5

(ArC), 149.5 and 158.6 (ArC and C=N).

Experimental 176

1, 5-0ipheny/-N, N'-bis[2-(2-pyridy/)ethy/]pentane-1, 5-diim ine 85

Attempted preparation

1 ,3-Dibenzoylpropane 83 (0.8 g, 3 mmol) and 2-(2-aminoethyl)pyridine 51 (0.76

mL, 6.3 mmol) were added to CHCI3 (10 mL), and the resulting mixture was boiled

under reflux for ca. 3.5 d. Completion of the reaction was confirmed by TLC. The

solvent was evaporated under reduced pressure, and the residual material was

chromatographed [flash chromatography on silica gel; elution with MeOH-benzene­

hexane (1 :4:1)] to yield a red-brown oil, 1H NMR analysis of which indicated the

absence of the required product.

J

Experimental 177

3.2.4 The Baylis-Hillman approach to ligand synthesis

3-Hydroxy-2-methy/ene-3-(2-pyridy/)propanenitrile 92149

A mixture of pyridine-2-carbaldehyde 91 (2.95 g, 28 mmol), acrylonitrile (1.54 g,

29 mmol) and DABCO (0.15 g, 1.3 mmol) in CHCls (2mL) was allowed to stand at

room temperature for 4d. The solvent was evaporated, and the residue

chromatographed (flash chromatography on silica gel; elution with EtOAc) to

afford, as colourless crystals, 3-hydroxy-2-methylene-3-(2-pyridyl)propanenitrile

92 (4.09 g, 91 %), mp 65-66°C (from hexane) (1it.149 66-6JOC); vmaithin film/cm-1)

3220 (OH) 2227 (CN) and 1600 (C=N); 0H(CDCls) 5.28 (2H, 2 x overlapping s,

CHOH and OH), 5.99 and 6.15 (2H, 2 x s, C=CH2), 7.23 (1 H, m, 5'-H), 7.38 (1 H,

d, 3'-H), 7.71 (1 H, m, 4'-H) and 8.49 (1 H, m, 6'-H).

3-Acetoxy-2-methyJene-3-(2-pyridyJ)propanenitrile 93149

A mixture of A~O (5mL) and 3-hydroxy-2-methylene-3-(2-pyridyl)propanenitrile 92

(1.0 g, 6.2 mmol) was heated at 100°C for 30 min. in a flask fitted with a reflux

condenser and drying tube. After cooling, the mixture was poured on to NaHCOs-

ice and, after stirring for 30 min., extracted with Et20 (2 x 100 mL). The combined

extracts were washed with aqueous NaHCOs (100 mL) and dried over anhydrous ~

MgS04. The solvent was removed under reduced pressure, and the resid,ue was

chromatographed [flash chromatography on silica gel; elution with EtOAc-hexane

(5:5)J to give, as a yellow oil, 3-acetoxy-2-methylene-3-(2-pyridyl)propanenitrile 93

(0.75 g, 60%); vrnax(thin film cm-1) 2226 (CN) and 1748 (CO); oH(400 MHZ; CDCls)

2.21 (3H, s, CH3CO), 6.14 and 6.17 (2H, 2 x s, C=CH2 ), 6.38 (1 H, s, CHOAc), 7.28

(1H, m, 5'-H), 7.48 (1H, d, 3'-H), 7.76 (1H, m, 4'-H), 8.60 (1H, m, 6'-H).

2-[(1-Piperidiny/)methyJ]-3-(2-pyridy/)prop-2-enenitri/e 96

Piperidine (0.64 g, 7.6 ml]]ol) was added to a solution of 3-acetoxy-2-methylene-3-

_. _ (2-pyridyl)propanenitrile 93 (1.5 g, 6.9 mmol) in THF (5mL), and the resulting

mixture was allowed to stand for 4 d at room temperature. Work-up, following the ~ -- .;

procedure described for preparing 3-(2-pyridy/)-1-[1-pyrro/idiny/)methyl]prop-2-

enenitri/e 97, gave, as a brown oil, 2-[(1-piperidiny/)methyl}-3-(2-pyridy/)-prop-2-

Experimental 178

enenitrile 96 (1.08 g, 69%) (Found: MH+, 228.1500. C14H17N3 requires MH,

228.1500); vmax<thin film/cm-1) 2217 (CN); oH(400 MHz; CDCI3) 1.40-1.47 (2H, m,

CH2), 1.55-1.65 (4H, m, 2 x CH2), 2.49 (4H, m, 2 x CH2), 3.29 (2H, d, CH2), 7.23

(1 H, CH=C), 7.28 (1 H, m, ArH), 7.73 (1 H, td, ArH), 7.86 (1 H, d, ArH), 8.65 (1 H, d,

ArH); od400 MHz; CDCI3) 24.1 (CH2), 25.9 (CH2), 54.2 (CH2), 62.9 (CH2), 112.9

(CH=C), 118.2 (CN), 123.5 (ArC), 124.1 (ArC), 136.7 (ArC),144.0 (CH=C), 149.9

(ArC) and 152.1 (ArC).

3-(2-Pyridy/)-2-[1-pyrro/idiny/)methy/]prop-2-enenitrile 97

A mixture of pyrrolidine (0.18 g, 2.5 mmol) and 3-acetoxy-2-methylene-3-(2-

pyridyl)propanenitrile 93 (0.50 g, 2.3 mmol) in THF (2mL) was allowed to stand at

room temperature for 4 d. The solvent was removed under reduced pressure, and

the residue was washed with NaOH (50 mL~ and extracted with CHCI3 (3 x 40 mL).

The combined extracts were dried with anhydrous MgS04, and the solvent was

evaporated under reduced pressure to give, as a brown oil, 3-(2-pyridy/)-2-[1-

pyrro/idiny/)methy/jprop-2-enenitri/e 94 (0.39 g, 80%) (Found: MH+, 214.1344.

C13H1sN3 requires MH, 214.1344); vmax(thin film/cm- 1) 2217 (CN); oH(400 MHz;

CDCI3) 1.81 (4H, s, 2 x CH2), 2.62 (4H, s, 2 x CH2), 3.44 (2H, s, CH2), 7.25-7.40

(2H, m, ArH and CH=C), 7.73 (1H, t, ArH), 7.82 (1H, d, ArH), 8.6~ (1H, d, ArH);

od 400 MHz; CDCI3) 23.7 (CH2), 53.8 (CH2), 59.8 (CH2), 113.4 (CH=C)! 118.2

(CN), 123.5 (ArC), 124.1 (ArC), 136.7 (ArC), 143.6 (CH=C), 149.8 (ArC) and 152.0

(ArC).

2, 10-Dicyano-1, 11-bis(2-pyridy/)-5, 8-diazaundeca-1, 11-diene 98

Attempted preparation

A mixture of 3-acetoxy-2-methy/ene-3-(2-pyridy/)propanenitri/e 93 (0.48 g, 7.3

mmol) and 1 ,3-diaminopropane (0.26 g, 3.7 mmol) in THF (6 mL) was allowed to

-. -stand at room temperature for 4 d. Completion of the reaction was confirmed by

TLC, and the THF was removed under reduced pressure. The residual dark-brown

oil was washed with aqueous NaHC03 (70 mL) and brine (60 mL). 1H NMR

Experimental 179

analysis of the fractions obtained by flash chromatography of the residue failed to

indicate the presence of the required product.

2,9-Bis(2-cyano-1-hydroxy-2-propenyl)-1, 10-phenanthroline 87

Attempted preparations

Method 1

A mixture of 1,10-phenanthroline-2,9-dicarbaldehyde 66 (1.0 g, 4.2 mmol),

acrylonitrile (0.46 g, 8.6 mmol) and DABCO (0.02 g, 0.4 mmol) in CHCI3 (8 mL)

was stirred at room temperature for 10 d. The CHCI3 was removed in vacuo and the

residue analysed by 1H NMR spectroscopy. The characteristic peaks for the

Baylis-Hillman product between 65.0-6.5 ppm were not observed in the spectrum.

Method 2

A mixture of acrylonitrile (0.94 g, 18 mmol), 1,10-phenanthroline-2,9-

dicarbaldehyde 66 (0.5 g, 2 mmol) and DABCO (0.02 g, 0.4 mmol) in CHCI3 (22

mL) was boiled under reflux for 40 min., followed by stirring at room temperature

for 7 d. After evaporating off the CHCI3, the residue was analysed by 1H-NMR

spectroscopy, but the characteristic peaks for the Baylis-Hillman product were not

observed.

Method 3

A mixture of acrylonitrile (0.94 g, 18 mmol), 1,1 0-phenanthr9Iine-2,9-

dicarbaldehyde 66 (0.5 g, 2 mmol) and DABCO (0.02 g, 0.4 mmol) in CHCI3 (22

mL) was stirred at room temperature for 7 d. After evaporating off the CHCI3, the

residue was analysed by 1 H NMR spectroscopy, but the characteristic peaks for

the Baylis-Hillman product were not observed.

-. -Method 4

1,10-Phenanthroline-2,9-diearbaldehyde 66 (~.O g, 4.2 mmol) was dissolved in hot

MeOH (110 mL) to give a clear yellow solution. DABCO (0.45 g, 9 mmol) and

Experimental 180

acrylonitrile (0.94 g, 18 mmol) were added to the hot solution, and the resulting

mixture was boiled under reflux for ca. 3.5 h. The MeOH was evaporated off under

reduced pressure and the residue analysed by 1 H NMR spectroscopy, which

indicated that the reaction had been unsuccessful.

J

Experimental 181

3.2.5 Macrocycle syntheses

The dibenzo macrocycle 99

A stirred mixture of 1,1 0-phenanthroline-2,9-dicarbaldehyde 66 (0.70 g, 3.0 mmol)

and 1 ,2-diaminobenzene (0.35 g, 3.0 mmol) in MeOH (85 mL) was boiled under

reflux for 5 min. The mixture was then stirred at room temperature for 12 h, during

which time, a yellow solid precipitated, which was filtered off to give, as a cream

powder, the dibenzo macrocycle 99 (0.56 g, 30%) mp 239-241°C (Found: MH+

616.2122. C4oH24Na requires MH, 616.2124); vmaiKBr/cm-1) 1619 (C=N).

The macrocycle 100

Attempted preparations

Method 1162

1 ,10-Phenanthroline-2,9-dicarbaldehyde 66 (0.70 g, 3.0 mmol) was dissolved in

MeOH (80 mL) by heating and stirring. The yellow solution was allowed to cool to

room temperature and was then added dropwise, during 40 min., to a solution of

1 ,2-diaminoethane MeOH (50 mL). The mixture was then stirred for 12 h at room

temperature. The MeOH was evaporated off in vacuo to afford a brown powder,

which was insoluble in DMSO, and clearly not the expected product.

Method 2

1,2-Diaminoethane (0.18 g, 3.0 mmol) was added to a hot solution ,af 1,10-

phenanthroline-2,9-dicarbaldehyde 66 (0.7 g, 3.0 mmol) in MeOH (70 ml), and the

resulting mixture was boiled under reflux for 2.5 h. The MeOH was removed under

reduced pressure to yield a brown powder, which was insoluble in DMSO, and was

not the expected product.

Experimental 182

3.2.6 Dendrimer-based ligands

Tris(2-cyanoethyl)nitromethane 102172

Acrylonitrile (3.98 g, 75 mmol) was added dropwise to a stirred solution of

nitromethane 101 (1.52 g, 24.9 mmol) and 40% aqueous Triton B (trimethylbenzyl­

ammonium hydroxide) (0.25 g) in dioxan (2.50 g, 62.5 mmol) over a period of ca.

25 min. while maintaining the temperature of the exothermic reaction at 25-30°C

by external cooling. The reaction mixture was then allowed to stand at room

temperature for 18 h. Dilute HCI was used to neutralise (pH 7) the reaction mixture,

which was then extracted with an equal volume of dichloroethane. The extract was

washed with H20 (2.5 mL), and the dichloroethane evaporated off under reduced

pressure. The crystalline product (2.15 g, 39%) was recrystallised from EtOH to

give, as colourless prisms, tris(2-cyanoethyl)nitromethane 102, mp 107-

109°C(lit.172 mp 11~C); oc(100 MHZ; MeOH-d4) 12.7 (CH2CH2CN), 31.4 (CH2CN),

92.3 (CN02) and 119.7 (CN).

4-[2-Carboxyethyl)]-4-nitroheptanedioic acid 103172

Tris(2-cyanoethyl)nitromethane 102 (1.5 g, 6.8 mmol) was dissolved in

concentrated HCI (6.5 mL), and the resulting solution boiled under reflux for 45

min., and then cooled to 5°C. The white solid which precipitated out was ,'filtered

off, washed with cold H20 (4 x 15 mL) and dried in vacuo to yield 4-[2-

carboxyethyl)]-4-nitroheptanedioic acid 103 (1.25 g, 66%), mp 186-18rC (lit. 172 mp

186°C); vrnax(KBr/cm1) 1721 (C=O), 2500-3500 (OH); od100 MHZ; MeOH-d4 ) 29.4

(CH2CH2C02H), 31.5 (CH2C02H), 93.9 (CN02) and 175.6 (C=O).

4-Nitro-4-(3-hydroxypropyl)-1, 7-heptanedio/1 04172

To a stirred solution of the triacid, 4-[2-carboxyethyl)]-4-nitroheptanedioic acid 103

(1.0 g, 3.6 mmol) in dry TRF (50 mL) at O°C, a BH3-THF solution (1.0 M; 11.9 mL,

-. - 11.9 mmol) was added dropwise. The temperature was allowed to rise to 25°C

once the white precipitatetlad formed. AfteFstirring for another 30 min., H20 was

slowly added until the solid had dissolved. Saturated NaHC03 was then added and

Experimental 183

the solvent was removed under reduced pressure. The residual oil was dried in

vacuo, and the resultant solid triturated with hot EtOH (3 x 60 mL), the ethanolic

solution being filtered in each case. The combined extracts were concentrated

under reduced pressure to give 4-nitro-4-(3-hydroxypropyl)-1 ,7 -heptanediol 104

(0.78 g, 93%); vmax(KBr/cm-1) 3040-3575 (OH); Od100 MHz; MeOH-d4) 26.6

(CH2CH20H), 31.7 (CH2CH2CH20H), 60.3 (CH20H) and 94.5 (CN02).

4-Amino-4-(3-h ydroxypropy/)-1, 7 -heptanedio/105172

A mixture of T-1 Raney Ni (ca. 3.0 g), nitro triol 104 (2.5 g, 10.6 mmol) and

absolute EtOH (100 mL) was shaken under H2 in a Parr hydrogenator at 3 atm for

3 d. The catalyst was cautiously removed by filtering the mixture through celite.

The solvent was then removed under reduced pressure, and the residual oil was

dried in vacuo to yield, as a semi-solid, 4-amino-4-(3-hydroxypropyl)-1, 7-

heptanediol105, shown by 1H NMR spectroscopy to be eventually pure, (1.67 g,

77%); OH(400 MHz; DMSO-d6) 1.20-1.26 (6H, m, CH2CH20H), 1.32-1.42 (6H, m,

CH2CH2C~OH), 3.35 (6H, t, CH2 0H), 4.61 (3H, br s, OH); ~(100 MHz; DMSO­

d6) 26.7 (CH2CH20H), 36.2 (CH2CH2CH20H), 52.1 (CNH2) and 61.6 (CH20H).

Preparation of the T-1 Raney nickel catalyst173

Raney nickel-aluminium alloy (50%; 20 g) was added in small portions durtng ca.

25 min. to a stirred aqueous NaOH solution (10%, 300 mL). The temperature was

maintained at 90-95°C during the addition. Once the addition was complete, the

mixture was stirred for 1 h, after which the nickel was allowed to settie, and the

supernatant solution was decanted. The residue was washed with water (5 x 100

mL) and then with EtOH (5 x 25 mL). The washing was done in such a way that the

catalyst was always covered with liquid. The catalyst was then kept under EtOH

and stored under refrigeration.

-. - The dendrimer-based ligand 112

To a hot solution of 1,1 0'"phenanthroline-2,9-dicarbaldehyde 66 (0.57 g, 2.4 mmol)

in absolute EtOH (60 mL) was added a solution of 4-amino-4-(3-hydroxypropyl)-

Experimental 184

1 ,7-heptanedioI105 (1.0 g, 4.9 mmol) in EtOH (30 mL), and the resulting mixture

was boiled under reflux for ca. 20 min. The completion of the reaction was

confirmed by IR spectroscopy. The solvent was removed under reduced pressure

to yield, as a brown oil, the dendrimer-based ligand 112 (1.4 g, 96%) (Found:

MNa+, 633.3627. C34HsoN4 Q, requires MNa, 633.3628); vmax (KBr/cm1 ) 1647 (C=N);

5H (400 MHz; DMSO-d6 ) 1.37 (24H, s, CH2CH2CH20H), 1.64 (12H, s, CH20H),

4.44 (6H, br s, OH), 8.09 (2H, s, CH=N), 8.36 (2H, d, ArH), 8.53 (2H, d, ArH) and

8.62 (2H, s, ArH); 5d100 MHz; DMSO-d6 ) 29.1 (CH2CH20H), 33.2

(CH2CH2CH20H), 63.5 (CH20H), 94.5, 119.6 (ArC), 127.3 (ArC), 129.34 (ArC),

137.1 (ArC), 144.8 (ArC) and 154.8 (ArC) and 157.5 (C=N).

2,9-Dimethyl-1, 10-phenanthroline-5,6-dione 109175,176 - attempted synthesis

To a mixture of neocuproine 64 (2.5 g, 12 mmol) and NaBr (2.5 g), was added an

ice cold mixture of H2S04 (30 mL) and HN03 (15 mL). The resulting mixture was

then boiled under reflux for 3 h. The hot solution was added to 400 mL of ice and

neutralised very cautiously with NaOH until neutral or slightly acidic. The resulting

mixture was extracted with CHCI3 and the extract dried with anhydrous MgS04.

The solvent was removed under reduced pressure, and analysis of the residue

revealed that the central ring in neocupoine was still intact.

J

The monoketone 111174 - attempted synthesis

To a boiling solution of neocuproine 64 (2.0 g) and KOH (1.0 g) in H20 (150 mL)

was added, in portions with stirring, a hot solution of KMn04 (5.0 g) in H20 (80

mL). The reaction mixture was then boiled under reflux for 1.5 h and filtered while

hot; the brown cake was washed with H20 (120 mL). The orange filtrate and

washings were extracted with CHCI3 (3 x 60 mL), and the combined extracts dried

with anhydrous MgS04. The CHCI3 was evaporated off under reduced pressure.

The residue was analysed by 1H NMR spectroscopy, which indicated that oxidation

_. _ of the central ring of neocuproine had not occurred. -

Experimental 185

3.3 SYNTHETIC PROCEDURES FOR THE PREPARATION OF COMPLEXES

3.3.1 Copper Complexes;

Preparation of tetrakis(acetonitrile)copper(l) hexafluorophosphate

Copper (I) oxide (4.0 g, 28 mmol) was dissolved in MeCN (80 ml) and stirred under

N2 in a closed flask fitted with a pressure-compensating dropping funnel. An aqueous

solution of HPF6 (60-65%; 10 mL, 113 mL) was added from the funnel, and the

mixture was stirred for at least 5 min. The mixture was then filtered and cooled to

-20°C to precipitate the crude product, which was recrystallised by redissolving in

MeCN (100 mL) and adding Et20 (100 mL), followed by cooling for approximately

6 h. The white crystalline complex [Cu(MeCN)4][PF6J (9,1 g, 44%) was filtered off,

dried under reduced pressure and stored at 4°C under N2 in a sealed container placed

in a desiccator.

Biphenyl and 1~ 10-phenanfhroline complexes

The copper complex 115

A clear, pink solution of 2,2'-bis{[2-(2-benzimidazolyl)ethylamino]tarbonyl}biphenyl I

55b (0.23 g, 0.44 mmol) in hot, dry, degassed DMF (20 mL) was added dropwise to

a stirred solution of [Cu(MeCN)4][PF61 (0.34 g, 0.92 mmol) in dry, degassed MeCN (10

mL). The resulting pale yellow-green solution gradually changed to a clear green

colour as the reaction mixture was stirred under N2 for 3.5 h. The addition of Et20

resulted in the precipitation of the copper complex 115 as a light-green, polymeric

solid (0.30 g), mp 198-202°C; vrnax(HCBD/ClTf1 ) 3237 (amide NH); vmax(nujollcm1) 1651

(CO) and 845 (PFs).

:t: See Table 8 (p.84) for microanalysis data.

Experimental 186

The copper complex 116

A solution of [Cu(MeCN)4][PFsl (0.34 g, 0.92 mmol) in dry, degassed MeCN (10 mL)

was added to a hot solution of 2,2'-bis{[2-(4-imidazolyl)ethylamino]carbonyl}biphenyl

55e (0.19 g, 0.44 mmol) in a mixture of dry, degassed DMF (10 mL) and dry,

degassed MeCN (20 mL). The reaction mixture turned blue and, upon addition of

Et20, the copper complex 116 precipitated out as a light-green solid (0.17g), mp 212-

214°C; vmaiHCBD/cm-1) 3242 (amide NH) and 3156 (imidazole NH); vmax(nujollcm-1

)

1652 (CO) and 846 (PFs).

Attempted preparation of the biphenyl dicopper complex 117

A colourless solution of [Cu(MeCN)4][PFs] (0.34 g, 0.92 mmol) in dry, degassed

MeCN (10 mL) was added to an almost colourless solution of 2,21-bis{[2-(2-

pyridyl)ethylaminojcarbonyl}biphenyf 55a (0.20 g, 0.44 mmol) in dry, degassed MeCN

(4 mL). No colour change was observed when the ligand solution was added

dropwise to the solution of the Cu(l) reagent, but a solid precipitated out of the

reaction mixture during stirring under N2 for 3.5 h. The solid was filtered off and Et20

was added to the filtrate to precipitate out more of the product. The microanalysis

results were not consistent with formation of the expected product. .

The copper complex 118a J

2,9-Bis{[2-(2-pyridyl)ethylamino]carbonyl}-1,1 O-phenanthroline 69a (0.20 g, 0.44

mmol) was dissolved in a mixture of hot, dry, degassed DMF (10 mL) and MeCN (15

mL). Dropwise addition of this solution to a colourless solution of [Cu(MeCN)4][PFsl (0.34 g, 0.92 mmol) in dry, degassed MeCN (10 mL) gave a dark-red mixture, the

colour of which changed to dirty green upon stirring under N2 for 6 h. The precipitated

solid was filtered off and Et20 was added to the filtrate to precipitate out the

remainder of the copper complex 118a, which was filtered off as a green powder

(0.27 g), mp >230°C; vm:(nujollcm-1) 1630 (CO) and 842 (PFs).

Experimental 187

The copper complex 118b

A brown solution of 2,9-bis{[2-(2-benzimidazolyl)ethylamino]carbonyl}-1, 1 O-phenan­

throline 6gb (0.25 g, 0.44mmol) in hot, dry, degassed DMF (20 mL) was added to a

colourless solution of [Cu(MeCN)4][PFsl (0.34 g, 0.92 mmol) in dry, degassed MeCN

(10 mL). The initially dark-red colour changed to green-brown and later to light-green

during stirring under N2 for 4.5 h. Et20 was added to precipitate the copper complex

118b as a green-brown solid (0.29 g), mp >230°C; vmax(HCBD/cm-1) 3227 (NH);

vmax(nujollcm-1) 1645 (CO) and 846 (PFs)'

The copper complex 118e

A hot, dark-brown solution of 2,9-bis{[2-(4-imidazolyl)ethylamino]carbonyl}-1, 10-

phenanthroline 6ge (0.19 g, 0.44 mmol) in dry, degassed DMF (20 mL) was added

dropwise to a colourless solution of [Cu(MeCN)4][PFs] (0.34 g, 0.92 mmol) in dry,

degassed MeCN (10 ml). The solution became dark red-brown in colour when stirred

under N2 for 5.5 h. Addition of E!:zO precipitated the copper complex 118e as a brown

solid (0.15 g), mp >230°C; vmax(HCBD/cm-1) 3157; vmax(nujollcm-1

) 1651(CO) and 845

(PFs)·

The copper complex 119

A solution of 2,9-bis{[2-(2-pyridylethylamino]methyl}-1,1 0-phenanthroline'68a (0.20

g, 0.44 mmol) in dry, degassed MeCN (50 mL) was added dropwise to a solution of

[Cu(MeCN)4JPFe (0.34 g, 0.44 mmol) in dry, degassed MeCN (10 mL). The reaction

mixture was stirred for 1.5 h under N2 before adding Et20 to precipitate, as a brown

solid, the copper complex 119 (0.38 g, 78%), mp >230°C; vmax(HCBD/cm1) 3273 (NH);

vmax(nujol/cm-1) 842 (PFs).

Experimental 188

Schiff base complexes and the macrocyclic complex

The copper complex 120

To a stirred solution of Cu(II)(N03h3H20 (0.47 g, 1.95 mmol) in MeCN (3.5 mL),

N,N'-bis[2-(2-pyridyl)ethyl]pentane-2,4-diimine 76 (0.30 g, 0.97 mmol) in MeCN (3.5

mL) was added dropwise. The colour of the reaction mixture changed from turquoise

to dark green during the addition. The reaction mixture was stirred for 3 d before Et20

was added and the copper complex 120 separated out as a green, hygroscopic oil

(0.68 g, 87%); vmaiKBr/cm-1) 1638 (C=N) and 1383 (N03).

The copper complex 121

A solution of 3-methyl-N,N'-bis[(2-(2-pyridyl)ethyl]pentane-2,4-diimine 77 (0.20 g,

0.62 mmol) in MeCN (2 mL) was added dropwise to a stirred solution of Cu(II)(N03h

.3H20 (0.30 g, 1.2 mmol) in MeCN (2.5 mL). The reaction mixture changed from

turquoise to dark green in colour during the addition. After stirring for 3 d, Et20 was

added to precipitate the crude complex, which was redissolved in MeCN. Evaporation

of the solvent under reduced pressure yielded, as a green, hygroscopic semi-solid,

the copper complex 121 (0.36 g, 58%); vmax(KBr/cm-1) 1637 (C=N), 1480, 1445, 1315,

1283 and 1013 (N03).

!

The copper complex 122 - attempted· synthesis

A solution of 3-ethyl-N,N'-bis-[2-(2-pyridyl)ethyl]pentane-2,4-diimine 78 (0.30 g, 0.89

mmol) in MeCN (3 mL) was added dropwise to a stirred solution of Cu(II)(N03h.3H20

(0.43 g, 1.78 mmol) in MeCN (2 mL). The colour changed from turquoise to dark blue

and, after 3 d, Et20 was added to affect separation of the copper complex.

Microanalysis of the resulting green hygroscopic semi-solid indicated that the

expected complex had not been obtained.

- .

Experimental 189

The copper complex 123

A solution of 3-benzyl-N,N'-bis[2-(2-pyridyl)ethyl]pentane-2,4-diimine 79 (0.30 g,

0.75 mmol) in MeCN (3 mL) was added dropwise to a stirred solution of

Cu(IJ)(N03h.3H20 (0.36 g, 1.5 mmol) in MeCN (3 mL). The colour of the reaction

mixture changed from turquoise to dark-green, and stirring was continued for 3 d

before Et20 was added to effect separation of the copper complex 123 as a green,

hygroscopic, semi-solid (0.41 g, 58%); vmaiKBr/cm-1) 1638 (C=N), 1483, 1383 and

1288 (N03).

The macrocyclic copper complex 124

A solution of macrocycle 99 (0.35 g, 1.1 mmol) in DMF (8 mL) was added to a stirred

solution of Cu(II)(N03h.3~O (0.52 g, 2.17 mmol) in DMF (4 mL). The colour changed

from turquoise to red-brown and, aftero~ stirring for 2.5 d, Et20 was added to

precipitate, as a dark-brown solid, the macrocyclic copper complex 124 (0.38 g, 35%),

mp >250°C; vmaiKBr/cm-1) 1653 (C=N) and 1383 (N03).

Attempted preparation of the copper complex 125

A solution of 1-phenyl-N,N'-bis[2-(2-pyridyl)ethyl]butane-1 ,3-diimine 81 (0.20 g, 0.54

mmol) in MeCN (3 mL) was added dropwise with stirring to a solution of

Cu(IJ)(N03h.3H20 (0.26 g, 1.1 mmol) in MeCN (2 mL). A colour chdnge from

turquoise to dark-green was observe(j during the addition and, after stirring for 3 d,

Et20 was added to precipitate the crude product which was redissolved in MeCN.

Evaporation of the solvent under reduced pressure gave a green, crystalline material,

microanalysis of which indicated that the desired complex had not been formed.

Attempted preparation of the copper complex 126

A solution of 1 ,3-diphenyl-N,N'-bis[2-(2-pyridyl)ethyl]propane-1 ,3-diimine 80 (0.40 g, ~

0.93 mmol) in MeCN (5 mL), was added dropw~se to a stirred solution of

Cu(lI)(N03h-3H20 (0.45 g, 1.9 mmol) in MeCN (3 mL). The reaction mixture was

Experimental 190

stirred for 1 d before adding Et20 to precipitate out the product, microanalysis of

which indicated that the expected complex had not been formed.

Attempted preparation of the dicoppercomplex 127

A solution of N,N'-bis[2-(2-pyridyl)ethyl]hexane-2,5-diimine 84 (0.30 g, 0.93 mmol) in

MeCN (5 mL) was added dropwise to a stirred solution of Cu(II)(N03)2.3H20 (0.45 g,

1.9 mmol) in MeCN (3 mL). After stirring the reaction mixture for 2 d, Et20 was added

to precipitate out the product, microanalysis of which indicated that the expected

complex had not been formed.

3.3.2 Cobalt Complexes§

The cobalt complex 133

To a stirred solution of CoCI2.6H20 (0.22 g, 0.92 mmol) in MeOH (5 mL), a solution

of 2,2'-bis{[2-(2-pyridyl)ethylamino]carbonyl}biphenyl 56a (0.20 g, 0.44 mmol) in

MeCN (5 mL) was added dropwise, and the resulting mixture was stirred for 48 h.

Et20 was then added to precipitate the cobalt complex 133 as a blue solid (0.21 g,

66%), mp 198-200°C; vmax(KBr/cm-1) 1622 (CO); vmax(nujollcm-1

) 3Q6 (Co-CI).

,I

The cobalt complex 134

A clear, colourless solution of 2,2'-bis{[2-(2-benzimidazolyl)ethylamino]carbonyl}­

biphenyl 55b (0.23 g, 0.44 mmol) in hot DMF (20 mL) was added to a stirred, blue

solution of CoCI2.6H20 (0.22 g, 0.92 mmol) in MeOH (5 mL) and stirred for 25 h. The

volume of the reaction mixture was reduced to less than half the original volume by

evaporating off the solvents under reduced pressure, and Et20 was added to

precipitate, as a blue solid, the cobalt complex 134 (0.24 g, 59%), mp >230°C;

§ See Table 12 (p.1 02) for microanalysis data.

Experimental 191

vmax(KBr/cm-1) 3185 (amide NH) and 1633 (CO); vmax(nujolfcm-1

) 295 (Co-CI).

The cobalt complex 135 - attempted preparation

A solution of 2,2'-{[2-(4-imidazolyl)ethylamino]carbonyl}biphenyl 55e (0.20 g, 0.46

mmol) in MeOH (10 mL) was added to a stirred solution of CoCI2.6H20 (0.23 g, 0.96

mmol) in MeOH (5 mL). After stirring for 48 h, during which time the colour changed

from light-blue to dark-blue, Et20 was added to precipitate the product. Microanalysis

of the resulting dark-blue hygroscopic semi-solid indicated that the desired complex

had not been obtained.

The cobalt complex 136

A solution of 1-[2-(2-pyridyl)ethyl]dibenzo[c,e]perhydroazepine 61 a (0.19 g, 0.44

mmol) in MeCN (5 mL) was added dropwisQ to a solution of CoCI2.6H20 (0.22 g, 0.92

mmol) in MeCN (10 mL), and the resulting mixture was stirred for 48 h. The MeCN

was allowed to evaporate off at room temperature, during which time, blue crystals

of the cobalt complex 136 precipitated and were filtered off (0.17 g, 89%), mp

>250°C; vmax(nujolfcm-1) 298 (Co-CI).

The cobalt complex 137a

A solution of 2,9-bis{[2-(2-pyridyl)ethylamino]carbonyl}-1, 1 0-phenanthroline169a (0.20

g, 0.44 mmol) in a mixture of hot MeOH (10 mL) and MeCN (5 mL) was added

dropwise to a purple solution of CoCI2.6~O (0.22 g, 0.92 mmoJ) in MeOH (4 mL). The

colour of the reaction mixture changed to dark-green during stirring for 22.5 h. The

reaction mixture was concentrated in vacuo to half the original volume. Et20 was

added to preCipitate, as a green solid, the cobalt complex 137a (0.26 g, 100%), mp

78°C; vmax(KBr/cm-1) 3274 (NH) and 1645 (CO); vmax(nujol/cm-1

) 303 (Co-CJ).

The cobalt complex 137b

-. - A solution of 2,9-bis{[2-(2-benzimidazolyl)ethylamino]carbonyl}-1, 1 O-phenanthroline

Experimental 192

69b (0.23 g, 0.40 mmol) in DMF (10mL) was added dropwise to a solution of

CoCI2.6H20 (0.20 g, 0.84 mmol) in DMF (5mL). The resulting mixture was stirred for

52 h, during which time, the colour changed from purple-blue to dark-green. The

volume of the reaction mixture was reduced to less than half by evaporating off the

solvent under reduced pressure. The subsequent addition of Et20 resulted in the

precipitation of the cobalt complex 137b as a green powder (0.32 g, 57%), mp

>230°C; vmaiKBr/cm1) 3251 (amide NH) and 1651 (CO); vmax(nujol/cm1) 311 (Co-CI).

The cobalt complex 137c

A solution of 2,9-bis{[2-(4-imidazolyl)ethylamino]carbonyl}-1, 1 O-phenanthroline 69c

(0.19 g, 0.44 mmol) in hot DMF (15 mL) was added to a stirred solution of CoCI2.6H20

(0.22 g, 0.92 mmol) in DMF (10 mL), resulting in a dark-green mixture. After stirring

for 41 h, the volume of the reaction mixture was reduced to less than half by

evaporating off the DMF under reduced pressure. This was followed by the addition

of Et20 which precipitated, as a dark-green solid, the cobalt complex 137c (0.31 g,

94%), mp 206-210°C; vmax{KBr/cm-1) 3259 (amide NH) and 3121 (imidazole NH);

vmax(nujol/cm-1) 299 (Co-CI).

The cobalt complex 138

A solution of 2,9-bis{[2-{2-pyridylethylamino]methyl}-1, 1 0-phenanthroline'68a (0.20

g, 0.44 mmol) in MeCN (10 mL) was added dropwise to a solution of CoCI2.6H20

(0.22 g, 0.92 mmol) in MeCN (10 mL). The colour changed from bright blue to green,

and, after stirring for 18 h the cobalt complex 138, precipitated ou~ as a green

powder (0.19 g, 49%), mp >230°C; vmax(KBr/cm-1) 3398 (NH); vmax(nujol/cm-1) 311

(Co-CI).

Experimental 193

3.3.3 Nickel Complexest ,lI

The nickel complex 140

A solution of 2,2'-bis{[2-(2-pyridyl)ethylamino]carbonyl}biphenyl 55a (0.20 g, 0.44

mmol) in MeOH (5 mL) was added to a stirred solution of NiCI2.6H20 (0.22 g, 0.92

mmol) in MeCN (5 mL). After stirring for 2.5 d, Et20 was added to precipitate, as a

light-green, hygroscopic solid, the nickel complex 140 (0.26 g, 72%), mp 48-49°C;

vrnaAKBr/cm-1) 3336 (amide NH) and 1619 (CO); vmaAnujol/cm-1

) 325 and 276 (Ni-CI).

The nickel complex 141

A hot solution of 2,2'-bis{[2-(2-benzimidazolyl)ethylamino]carbonyl}biphenyI 55b (0.23

g, 0.44 mmol) was added dropwise to a stirred solution of NiCI2.6H20 (0.22 g, 0.92

mmol) in DMF (7 mL). The reaction mixture was stirred for 57 h before Et20 was

added to precipitate, as a light-green solid, the nickel complex 141 (0.29 g, 53%), mp

>250°C; Vmax (KBr/cm-1) 3183 (amide NH) and 1650 (CO); vmaAnujol/cm1

) 387 (Ni-CI)

The nickel complex 142

A solution of 2,2'-bis{[2-(4-imidazolyl)ethylamino]carbonyl}biphenyl 55c (0.15 g, 0.35

mmol) in DMF (10 mL) was added dropwise to a stirred soluti6n of NiCI2.6H20

(0.17 g, 0.72 mmol) in MeOH (5 mL). Stirring was continued for 55 h befdre adding

Et20 to precipitate, as a hygroscopic dark blue-green solid, the nickel complex 142,

(0.10 g, 27%), mp 204-205°C; vmax(KBr/cm-1) 3248 (amide NH), 3143 (imidazole NH)

and 1653 (CO); vmax(nujol/cm-1) 378 (Ni-CI).

The nickel complex 143

To a stirred solution of NiCI2.6H20 (0.22 g, 0.92 mmol) in MeOH (4 mL), a solution of

1-[2-(2-pyridyl)ethyl]dibenz[c,e]perhydroazepine 61 a (0.19 g, 0.44 mmol) in MeCN

_ . t While 1H and 13C NMR spectra were obtained for each of these-complexes, the complexity of - their spectra (see Discussion 2.3.4.2.2 p.118) precludes systematic peak assignment. ~ See Table 15 (p.11S) for microanalysis data.

Experimental 194

(10 mL) was added dropwise, and stirring was continued for 19 h. The solvent was

allowed to evaporate off at room temperature, during which time, purple crystals

precipitated from the reaction mixture and were filtered off to give the nickel complex

143 (0.13 g, 65%), mp >250°C; vmax(nujollcm-1) 329 and 289 (Ni-CI).

The nickel complex 144a

To a stirred solution of NiCI2.6~O (0.20 g, 0.92 mmol) in MeOH (5 mL), a solution of

2,9-bis{[2-(2-pyridyl)ethylamino]carbonyl}-1, 1 O-phenanthroline 69a (0.20 g, 0.44

mmol) in DMF (5 mL) was added dropwise. Although no immediate colour change

was observed, a suspension formed during stirring for 48 h. Et20 was then added to

the reaction mixture to precipitate, as a light green solid, the nickel complex 144a

(0.24 g, 71 %), mp 249-252°C; vmax (KBr/cm-1) 3223 (amide NH) and 1653 (CO); vmax

(nujollcm-1) 378 and 253 (Ni-CI).

The nickel complex 144b

A solution of 2,9-bis{[2-(2-benzimidazolyl)ethylamino]carbonyl}-1, 1 O-phenanthroline

69b (0.22 g, 0.44mmol) in DMF (15 mL) was addded dropwise to a solution of

NiCI2.6H20 (0.20 g, 0.92 mmol) in DMF (7 mL). After stirring for 52 h E~O was added

to precipitate, as a brown, hygroscopic solid, the nickel complex 14t1-b (0.31 g, 66%),

mp >250°C; Vmax (KBr/cm-1) 3245 (amide NH) and 1653 (CO); vmax(nujollcm') 384 and

283 (Ni-CI).

The nickel complex 144c

A solution of 2,9-bis{[2-(4-imidazolyl)ethylamino]carbonyl}-1, 1 O-phenanthroline 69c

(0.20 g, 0.44 mmol) in DMF (6.5 mL) was added dropwise to a stirred solution of

NiCI2.6H20 (0.21 g, 0.88 mmol) in DMF (5 mL). The resulting suspension was stirred

for 20.5 h before adding Et20 to precipitate, as a light-green solid, the nickel complex

144c (0.31 g, 94%), mp >250°C; vmax (KBr/cm-1) 323f? (amide NH) and 1649 (CO);

vmax(nujollcm-1) 373 and 310 (Ni-CI).

Experimental 195

The nickel complex 145

A solution of 2,9-bis{[2-(2-pyridylethylamino]methyl}-1,1 O-phenanthroline 6Sa (0.25

g, 0.56 mmol) in MeOH (5 mL) was added dropwise to a solution of NiCI2.6H20 (0.26

g, 1.11 mmol) in MeOH (5 mL). The colour changed from light-green to dark-green

and, after stirring for 48 h, Et20 was added to precipitate, as a light-green solid, the

nickel complex 145 (0.30 g, 70%), mp 244-246°C; vmaAKBr/cm-1) 3367 (NH);

vmax(nujollcm-1) 352 and 278 (Ni-CI).

3.3.4 Platinum Complexes tt, t

The platinum complex 146

A solution of 2,2'-bis{[2-(2-pyridyl)ethylamino]carbonyl}biphenyl 56a (0.12 g, 0.26

mmol) in MeCN (5 mL) was added dropwise to a stirred solution of K2[PtCI4] (0.24 g,

0.57 mmol) in H20 (3 mL). After stirring for 51 h, the solvents MeCN and H20 were

evaporated, almost to dryness, under reduced pressure, before adding H20 to

precipitate out, as a yellow solid, the platinum complex 146 (0.16 g, 55%), mp 209-

214°C; vmax(HCBD/cm-1) 3252 (amide NH); vmax(nujol/cm-1

) 1634 (CO) and, 342 and

315 (Pt-CI). J

The platinum complex 147 - Attempted synthesis

To a solution of ~[PtCI4] (OAO g, 0.76 mmol) in DMF (84 mL), a solution of 2,2'-bis{[2-

(2-benzimidazolyl)ethylamino]carbonyl}biphenyl 56b (0.20 g, 0.38 mmol) in DMF (6

mL) was added dropwise while stirring. After stirring for 7 d, the volume of DMF was

evaporated to less than half the original volume before adding Et20 to precipitate out

a cream powder, microanalysis of which indicated that the desired complex 147 had

not been obtained.

_. _ tt While the 1H and 13e NMR spectra were obtained for each of these complexes, the complexity of their spectra (see Discussion 2.3.4.2.2 p.130) precludes systematic peak assignment.

t See Table 18 (p.122) for rnifroanalysis data.

- .

Experimental 196

The platinum complex 148

A solution of 2,2'-bis{[2-(4-imidazolyl)ethylamino]carbonyl}biphenyI56c (0.11 g, 0.26

mmol) in DMF (6 mL) was added dropwise to a stirred solution of K2[PtCI4] (0.24 g,

0.57 mmol) in H20 (3 mL). After stirring for 52.5 h, the solvents were evaporated

under reduced pressure to afford, as a light-brown solid, the platinum complex 148

(0.16 g, 62%), mp >230°C; vmax(nujol/cm1) 3218 (amide NH), 3120 (imidazole NH) and

1627 (CO); vmaAnujol!cm-1) 335 (Pt-CI).

The platinum complex 149

The dropwise addition of a solution of 2,9-bis{[2-(2-pyridyl)ethylamino]carbonyl}-1, 10-

phenanthroline 69a (0.11 g, 0.23 mmol) in DMF (4 mL) to a stirred solution of

K2[PtCI4] (0.21 g, 0.50 mmol) in H20 (4 mL) was accompanied by the formation of a

suspension. After stirring for 65 h, the precipitate was filtered off to afford, as an

orange-brown powder, the platinum complex 149 (0.22 g, 100%), mp >230°C;

vmax(nujol!cm-1) 1660 (CO); vmax(nujollcm-1

) 329 (Pt-CI).

The platinum complex 150 - Attempted synthesis

A solution of 2,9-bis{[2-(2-benzimidazolyl)ethylamino]carbonyl}-1, 1 O-phenanthroline

69b (0.15 g, 0.27 mmol) in DMF (10 mL) was added dropwise to a ~tirred solution of

K2[PtCI4] (0.23 g, 0.54 mmol) in DMF (20 mL) and H2 0 (8 mL). The resultirig mixture

was stirred for 6 d before concentrating the mixture in vacuo to less than half the

original volume. Et20 was then added to precipitate a peach-coloured solid,

microanalysis of which indicated that the expected complex 150 had not been

obtained.

The platinum complex 151

A solution of 2,9-bis{[2-(4..:imidazolyl)ethylamino]carbonyl}-1, 1 O-phenanthroline 69c

(0.20 g, 0.44 mmol) in DMF (5 mL) was added dropwise to a stirred solution of

K2[PtCI2] (0.21 g, 0.50 mmol) in DMF (20 mL) and H20 (9 mL). After stirring for 6 d,

Experimental 197

the mixture was concentrated to less than half the original volume under reduced

pressure. The addition of Et20 resulted in precipitation of the platinum complex 151

as an orange-yellow solid (0.43 g, 100%), mp >250°C); vmax(KBr/cm-1) 3259 (amide

NH), 3135 (imidazole NH) and 1653 (CO); vmax(nujollcm-1) 321 (Pt-CI).

The platinum complex 152

To a solution of ~[PtCI4] (0.21 g, 0.50 mmol) in ~O (3 mL) and HCI (2.5 mL, 2M) was

added dropwise 2,9-bis{[2-(2-pyridylethylamino]methyl}-1, 1 O-phenanthroline S8a

(0.11 g, 0.23 mmol) in a mixture of DMF (4 mL) and MeCN (26 mL). The colour

changed from orange-red to dark-brown and a suspension formed during stirring for

5 d. After this time, the mixture was filtered and the filtrate evaporated, almost to

dryness, under reduced pressure. Addition of H20 to the residue afforded, as a brown

solid, the platinum complex 152 (0.25 g, 100%), mp >230°C; vmax(nujollcm-1) 3480

(NH) and 318 (Pt-CI).

J

Experimental 198

3.4 PROCEDURE FOR COMPUTER MODELLING OF COMPLEXES

Computer modelling was conducted using the MSI Cerius2 version 3.2 modelling

package on a Silicon Graphics 0 2 platform. The sequence followed for the

computer modelling runs for both ligands and complexes is outlined in Figure 63,

and was aimed at establishing the global minimum in each case.

Structure

building

Dynamics run Dynamics anneal cycles • at 300 K --... at 500 K

Minimisation .04-- Dynamics anneal trajectory at 300 K

Figure 63: The procedure followed for the modelling of the ligands and complexes.

This sequence involved starting with structure building followed by a dynamics run

of 5000 steps at 300 K. This was then followed by 250 dynamics anneal cycles

during which the temperature was raised to 500 K followed by a drnamics anneal

trajectory involving cooling to 300 K. The lowest energy structure was then, energy­

minimised. Minimisations were consi~ered to converge when the energy gradient

was less than 0.01 kcalmol-1. The ligands were first modelled and then the metals

were inserted and the modelling process repeated for the complex. Dioxygen was

inserted once the minimised model of the complex was obtained, and the

modelling process once again repeated for the copper-dioxygen complex. The

potential energies and the Cu-Cu distances of these minimised models were then

recorded for selected complexes. A universal force field was used for modelling all

the complexes except oepper.

Experimental 199

3.5 ELECTROCHEMICAL STUDIES: CYCLIC VOL TAMMETRY

Electrochemical data were collected with a Bio Analytical Systems (BAS) CV-50

voltammograph. The measurements were carried out under a nitrogen atmosphere

using freshly distilled, dried, degassed solvents. A glass carbon electrode

(diameter = 3.0 mm) was used as the working electrode and a platinum wire as the

counter electrode. A non-aqueous reference electrode was employed. The

electrode solution for the non-aqueous reference electrode was prepared by

dissolving 0.01 M AgN03 in 0.1 M TEAP (tetraethylammonium perchlorate) in

MeCN or DMF, resulting in a Ag/Ag+ (TEAP/MeCN or TEAP/DMF) electrode.

TEAP was employed as an electrolyte. The same reference electrode was used

for both MeCN and DMF.

3.4.1 Copper complexes

The conditions used for the analysis of the copper complexes are summarised in

Table 26. The resulting cyclic voltammograms are illustrated in Figures 65-67.

Cyclic voltammograms were also obtained for the ligands 55b and 6gb (Figure 64).

Table 26: Conditions used to measure the cyclic voltammogram~ of the copper complexes. J

Complex Solvent Scanning Scanning speed

range (mV) (mV.s·1)

115 DMF 1300 to -1300 1000

116 DMF 1550 to -1550 500

118a DMF 1580 to -1580 200

118b DMF 2200 to -2200 1000

118c DMF 2000 to -2050 500 r-

11 9 DMF 1200 to -1200 200

120 MeCN 2200 to -2200 300

121 - - MeCN 22QO to -2200 500

123 MeCN 2250 to -2250 100

127 DMF 1650 to -1650 500

Experimental 200

+16 1 ! ! ! ! ! ! ! ! I I I I

1 +151 I I I I I I I ! I I I

I

J I

1 ....-,L

1 I I I

OJ r +51 L I

~ ~ ~ ~ J I

I

I I 1 I

<C -101 I r <C -51 r ::::> I ::::> I -E I

~ -E 1 t ~ i ~ ::::> I l ::::>

-151 l 0

-321

0

I f I 1 \ 1 I

~ I

-48~ I

-25~ I ~ '-I

I I I I i

~J r 1 r I ! I I I I I I I I I I I I -351 I I I I I I I I I I I I I 1

+1.4 +1.0 +0.0 +0.2 -0.2 -{).6 -1.0 -1.4 +1.4 +1.0 +0.0 +0.2 -{).2 -0.0 -10 -1.4

Potential,V Potential,V

(a) (b)

Figure 64: Cyclic voltammograms of: (a) the biphenyl ligand 55b and (b) the 1,10-phenanthroline ligand 6gb.

+80.00j i ~ +60.00 L

+0000 • ._.//~.~) f +20.00 ~ ./ ~ i .:oO::jl ;i L

<3 -40.00 .~ ,~.~ ~

I' r ~::::j _I! L~ -100.01 I

-120.0 i I I ! + 1.40 .+-1 ,':-C -1-0.50 +C.2~ -C.20 -0.60 -l.CJ -1.4;

Potentiol/V

Figure 65: Cyclic voltammogram of the oiphenyl copper complex 115.

Experimental

:1··········~ 1 r

c ~ ~l :3_ J' "E

~ I I o ~i ~

c :3

-E' ~ ::J

0

'V -40 / ,

+1.6 +1.2

+80 I J I I

+4°1

j

o~ I I 1

-401

i I

-80~

-120 i I

~ I r ~ I

I I I I I I I

+0.8 +0.4 ° -D.4 -D.S -1.2 -1.6

I I

Potential,V

(a)

I ~

I r ~ I r r I , e , I . r , I L I

I r

I I I I I,

+2.0 +1.5 +1.0 +0.5 -D.S -1.0 -1.5 -2.0 -2.5

Potential,V

(C)

c :3

-E' ~ ::.

0

+48

1

~ ,

+32~ I

-1 I

+161

I +1

I ' I o -1

Potential,V

(b)

, ,

o~ -16~

-32/ , , , , +1.2 +0.8 +0.4 a -D.4

Potential,V

(d)

, ,

, -D8

I -2

,

~ I r ~ l

.I I I ~ I r ~ I I ,

-1.2

Figure 66: Cyclic voltammograms of the 1,1 O-phenanthroline complex-es: (a) 118a; (b) 118b; (c) 118c; and (d) 119.

201

Potential ,V

(a)

\ 1 \

, o

Potential,V

(c)

I -1

Experimental

+800+-

J --"--, -1'----'-' -----'---'---'-----L----L-I ~I --tl

I I +4001 r

~ ~ J ~

~ 1 I ~ I r 8 AOOJ l .- I I

i I 1 I

-800~ ~ I I 1 I

-1200 I I I I I I I I ~ ~ 0 ~ ~

Potential,V

(b)

I +45Oj I I I I I ' I I I ' I , I

t ~J t

1 1 ~ ~ <~~ ~ l ~- J l 1 ~ I 1

r <..) -45Oi r ~ ~ ,I ~

Ir ___ I 1 -I~l r

t 1 I I I -1050 I . I . I . I . I . I ' I . I . I I

-2 +1.8 +1.4 +1.0 +0.6 +0.2 -0.2 -0.6 -1.0' -1.4 -1.8

Potential,V

(d)

Figure 67: Cyclic voltammograms obtained for the Schiff base complexes

202

and the macrocycle complex: (a) 120; (b) 121; (c) 123; and (d) 124_

Experimental 203

3.4.2 Cobalt Complexes

The conditions used for the analysis of the cobalt complexes are summarised in

Table 27.

Table 27: Conditions used to measure the cyclic voltammograms of the cobalt complexes.

Complex Solvent Scanning Scanning

range (mV) speed (mV.s·1)

133 DMF 1600 to -1600 500

134 DMF 1250 to -1250 300

137a DMF 1600 to -1600 400

137b DMF 1250 to -1250 300

137c DMF 1300 to -1300 400

138 DMF 1250 to -1250 300

.1

'-

Experimental 204

3.4.3 Nickel Complexes

The conditions used for the analysis of the nickel complexes are summarised in

Table 28.

Table 28: Conditions used to measure the cyclic voltammograms of the nickel complexes.

Complex Solvent Scanning Scanning

range (mV) speed (mV.s-1)

140 DMF 1550 to -1550 300

141 DMF 1550 to -1550 500

142 DMF 1550 to -1550 300

144a DMF ''1550 to -1550 300

144b DMF 1550 to -1550 700

144c DMF 1550 to -1550 500

145 DMF 1550 to -1550 - 500

,I

Experimental 205

3.6 METHOD FOR THE EVALUATION OF THE CATALYTIC ACTIVITY OF

THE COPPER COMPLEXES

The substrates DTBP 153 and DTBC 154 were added to separate suspensions of

the complexes in dry DMF (3 mL) or CH2CI2 (3 mL) in molar ratios of 100: 1

(substrate: complex). The resulting mixtures were aerated by vigorously stirring

at room temperature to provide the oxygen necessary for reaction. Generally the

reactions conducted in CH2CI2 were stirred for ca. 3-4 d, reactions in DMF for ca.

5 d; the reactions with complexes 115 and 118a, however, showed biomimetic

activity after stirring for 24 h. The reactions in DMF, which displayed no catalytic

activity, were repeated in the presence of triethylamine (30 IJL per reaction) to

inhibit decomposition of the complex. 138 After stirring, the reaction mixtures were

concentrated to dryness in vacuo, and tile solid residues analysed by 1H NMR

spectroscopy. The extent of conversion to the products was determined by

comparing the integrals of signals corresponding to the individual products and

unreacted substrate (see Tables 29 and 30).

Table 29: Data for the catalytic oxidation of DTBP 153 to the coupled product 156.

Complex Integral ratio % conversion

116 10.33:1 8.8

118a 6:1 14.3

118b 6.67:1 13.0

Table 30: Data for the catalytic oxidation of DTBC 101 to DTBQ 102.

Complex Integral ratio % conversion

115 15: 1 6.3 ~

116 13:2 13.3

118a 12.7: 1 7.3 - - <

118b 5:3 38.0

118c 1: 1.19 54.3

J

206

4 REFERENCES

1. R. J. P. Williams, Endeavour, 1967,26,96.

2. A. S. Milvan and M. Cohn, Adv. Enzymol., 1970, 33,1.

3. W. N. Lipscomb and N. Strater, Chem. Rev., 1996,96,2375

4. G. C. Dismukes, Chem. Reviews, 1996, 96, 2909.

5. R. W. Hay, Bio-inorganic Chemistry, Ellis Horwood Ltd., Chichester, 1987,

p.102.

6. D. E. Wilcox, Chem. Rev., 1996,96,2435.

7. S.J. Lippard and J. M. Berg, Principles of Bioinorganic Chemistry,

University Science Books, Mill Valley, California, 1994, p.31.

8. D. W. Christianson, Acc. Chem. Res., 1989,22,62 ..

9. K.D. Karlin, Science, 1993,261,701.

10. C. A. Mcauliffe, Techniques and Topics in Bioinorganic Chemistry, The

McMillan Press Ltd, London, 1975, Pp.3-5.

11. H. S. Mason, W. B. Fowlks and E. W. Peterson, J. Am. Chem. Soc., 1955,77,

2914; E. Bourquelot and G. Bertrand, Compt. Rend. Soc. Bio!., 1895,47,582.

12. J. N. Rodriguez-Lopez, J. Tudela, R. Varon, F. Garcia-Carmona, F. Garcia­, Carnovas, J. BioI. Chem., 1992,267,3801.

.1

13. T. Tsukihara, H. Aoyama, E. Yamashita, T. Tomizaki, K. Shinwaza-Itoh, R.

Nakashima, R, Yaono, S. Yoshika-wa, Science, 1995, 269, 1069.

14. K. D. Karlin and Y. Gultneh, J. Chem. Ed., 185,62,983.

15. P. A. Vigato, S. Tamburini and D. E. Fenton, Coord. Chem. Rev., 1990,

106,25.

16. G. L. Woolery, L. Powers, M. Winkler, E. I. Solomon, K. Lerch and T. G.

Spiro, Biochim. Biophys. Acta, 1984, 788, 155.

17. N. Makino and H. S. Mason, BioI. Chem., 1973,248,5731.

-. _ 18. D. E. Wilcox, A. G. Porras, Y. T. Hwang, K. Lerch, -M. Winkler and E. L

Solomon, J. Am. Chf?l7J. Soc., 1985, 10~, 4015.

19. T. B. Freedman, J. S. Loehr and T. M. Loehr, J. Am. Chem. Soc, 1986,98,

2809.

207

20. N. Kitajima, Adv. Inorganic Chem., 1992, 39, 1.

21. P. K. Ross and E. I. Solomon, J. Am. Chem. Soc., 1990, 112,5871.

22. P. K. Ross and E. I. Solomon, J. Am. Chem. Soc., 1991, 113, 3246.

23. M. J. Baldwin, D. E. Root, J. E. Pate, K. Fujisawa, N. Kitajima and E. I.

Solomon, J. Am. Chem. Soc., 1992, 114, 10421.

24. S. Mahapatra, J. A. Halfen, E. C. Wilkinson, G. Pan, C. J. Cramer, L. Que

and W. B. Tolman, J. Am. Chem. Soc, 1995, 117,8865.

25. J. A. Halfen, S. Mahapatra, E. C. Wilkinson, S. Kaderli, V. G. Y. L. Que, A.

D. Zuberbuhler and W. B. Tolman, Science, 1996,271,1397.

26. C. H. Langford and H. B. Gray, Ligand Substitution Processes, Benjamin,

Reading MA, 1966.

27. K. Lerch and M. Huber, J. Inorg. Biochem., 1986,26,213.

28. K. Lerch in Enzymatic Browning and Its Prevention, eds. C. Y. Lee and J. R.

Whitaker, ACS, Washington, DC, 1995.

29. W. H. Lang and K. E. Holde, Proc. Nat/. A cad. Sci. U. S.A., 1991, 88, 244.

30. E. Pfiffner and K. Lerch, Biochemistry, 1981, 20, 6029.

31. M. P. Jackman, A. Hajnal and K. Lerch, Biochem. J., 1991,274,707.

32. C. Gielens, N. D. Geest, X. -Q. Xin and G. Preaux, Arch. Int. Physiol.

Biochim. Biophys., 1991,102, B11. J

33. N. Ito, S. E. V. Phillips, C. Stevens, Z. B. Ogel, M. J. McPherson, J. N. Keen

and P. F. Knowles, Nature, 1991, '350,87.

34. A J. M. S. Uiterkamp, FEBS Lett., 1972, 20, 93.

35. A J. M. S. Uiterkamp, H. v. d. Deen, J. Berendsen and J. F. Boas, Biochem.

Biophys. Res. Commun., 1974,372,407.

36. N. C. Eickman, R. S. Himmelwright and E. I. Solomon, Proc. Nat!. Acad. Sci.

U.S.A., 1979,76,2094.

37. A. F. Hepp, R. S. HirTlmelwright, N. C. Eickman and E. I. Solomon, Biochem.

Biophys. Res. Commun., 1979,89, 1050.

38. R. S. Himmelwright, N. C. Eickman and E. I. Solomon, Biochem. Biophys.

Res. Commun., 1978,81,237.

39. R. S. Himmelwright, N. C. Eickman and E. I. Solomon, Biochem. Biophys.

208

Res. Commun., 1978, 84, 300.

40. R. S. Himmelwright, N. C. Eickman and E. I. Solomon, Biochem. Biophys.

Res. Commun., 1979, 86, 628.

41. R. S. Himmelwright, N. C. Eickman, C. D. Lubien and E. I. Solomon,

Biochem. Biophys. Res. Commun., 1980, 102,5378.

42. R. S. Himmelwright, N. C. Eickman, C. D. Lubien, K. Lerch and E. I.

Solomon, J. Am. Chem. Soc., 1980, 102,7339.

43. R. L. Jolley, L. H. Evans and H. S. Mason, Biochem. Biophys. Res.

Commun., 1972, 46, 878.

44. R. L. Jolley, L. H. Evans, N. Makino and H. S. Mason, J. BioI. Chem., 1974,

249,335.

45. A J. M. S. Uiterkamp and H. S. Mason, Proc. Nat!. A cad. Sci. U.S.A., 1973,

70,993.

46. N. C. Eickman, E. I. Solomon, J. A. Larrabee, T. G. Spiro and K. Lerch, J.

Am. Chem. Soc., 1978, 100, 6529

47. K. A. Magnus, B. Hazes, H. Ton-That, C. Bonaventura, J. Bonaventura and

W. G. J. Hal, Proteins, 1994, 19,302.

48. S. H. Pomerantz, J. BioI. Chem., 1960,241,161. ~

49. H. W. Duckworth and J. Coleman, J. BioI. Chem., 1970,245, 1613. J

50. E. I. Solomon, U. M. Sundaram and T. E. Machonkin, Chem. Rev., 1996, 96,

2563.

51. S. C. Atlow, L. Bonadonna-Aparo and A. M. Klibanov, Biotech. Bioeng.,

1984, 26, 599.

52. S. Wada, H. Ichikawa and K. Tatsumi, Biotech. Bioeng., 1995, 45, 304.

53. S. Wada, H. Ichikawa and K. Tatsumi , Biotech. Bioeng., 1993, 42, 854.

54. J. G. Schiller and C. C. Liu, Biotech. Bioeng., 1993,118,979.

55. J. G. Schiller, A. K. Cben and C. C. Liu, Anal. Biochem., 1978,85,25.

_. _56. K. G. Kjellen and H. Y. Neujahr, Biotech. Bioeng., 1-980, 22, 299.

57. Z. Tyeklarand K. D. Karlin, Acc. Chem. Res.,1989, 22, 241. ~ -- >:

58. T. N. Sorrell, Tetrahedron, 1989, 45, 3.

59. K. D. Karlin, Science, 1993,261, 701.

209

60. K. D. Karlin and Y. Gultneh, Prog. Inorg. Chem., 1987,35,219.

61. L. Casella, O. Carugo, M. Gullotti, S. Garfani and P. Zanello, Inorg. Chem.,

993, 32, 2056.

62. P. V. Bernhart and P. Comba, Inorg. Chem., 1992,31,2638.

63. T. Pandiyan, M. Palandiandavar, M. Lakshminarayanan and H. Manohar, J.

Chem. Soc. Dalton Trans., 1992,3377.

64. M. R. Malchowski, L. J. Tomlinson, M. G. Davidson and M. J. Hall, J. Coord.

Chem., 1992, 25, 171.

65. K. D. Karlin, P. L. Dahlstrom, S. N. Cozzette, P. M. Scensny and J. Zubieta,

J. Chem. Soc., Chem. Commun., 1981, 881.

66. G. S. Patterson and R. H. Holm, Bioinorg. Chem., 1975,257,4.

67. K. D. Karlin and J. K. Yandell, Inorg. Chem., 1984,23, 1184.

68. A W. Addison in Copper Coordination~-Chemistry: Biochemical and

Inorganic Perspectives, eds. K. D. Karlin and J. Zubieta, Adenine,

Guilderland, New York, 1983, p. 109-128.

69. R. J. Kassner, Proc. Nat/, Acad. Sci., 1972,69,2263.

70. K. D. Karlin, Prog.lnorg. Chem., 1987,35,219.

71. G. L. Woolery, L. Powers, M. Winkler, E. I. Solomon and T. G. Spiro, J. Am.

Chem. Soc., 1984, 106,86. !

72. T. B. Freedman, J. S. Loehr and T. M. Loehr, J. Am. Chem. Soc., 1976,98,

2809.

73. J. A Larrabee and T. G. Spiro, J. Am. Chem. Soc., 1980, 102,4217.

74. J. A. Larrabee and T. G. Spiro, J. Am. Chem. Soc.,1980, 102,4217.

75. M. K. Jacobsohn, V. C. Dobre, C. Branam and G. M. Jacobsohn, J. Steroid

Biochem., 1988,31,377.

76. E. I. Solomon, Pure Applied Chem., 1983, 55, 1069.

77. W. P. Gaykema, A. \[olbeda and W. G. J. Hoi, J. Mo/' Bio!., 1985, 187,255.

_. _ 78. D. E. Fenton in Biocoordination Chemistry, Oxford-University Press, New

York, 1995, p.29.

79. N. Kitajima in Bioinorganic Chemistry of Copper, eds K. D. Karlin and Z.

Tyeklar, Chapman and Hall, New York, 1993, p.251.

80. K. A. Magnus, H. Ton-That and J. E. Carpenter, Chem. Rev., 1994,94,

727.

81. E. Spodine and J. Manzur, Coord. Chern. Rev., 1992, 119, 171.

210

82. M. G. Simmons and L. G. Wilson, J. Chem. Soc., Chern. Commun.,1988,

151.

83. M. G. Simmons, C. L. Merrill, L. G. Wilson, L. A. Bottomley and K. M.

Kadish, J. Chem. Soc., Dalton Trans.,1980, 1827.

84. J. D. Korp, I. Bernal, C. L. Merrill and L. J. Wilson, J. Chern. Soc., Dalton

Trans., 1981, 1951.

85. C. L. Merrill, L. J. Wilson, T. J. Thamann, T. M. Loehr, N. S. Ferris and W.

H. Woodruff, J. Chern. Soc., Dalton Trans., 1984, 2207.

86. L. Casella, M. E. Silver and J. A. Ibers, fnorg. Chem., 1984,23,1409.

87. R. R. Jacobson, Z. Tyeklar, A. Faraoq~' K. O. Karlin, S. Liu and J. Zubieta, J.

Am. Chem. Soc., 1988, 110,3690.

88. N. Wei, N. N. Murthy, Z. Tyeklar and K. O. Karlin, fnorg.Chem., 1994,33,

1177.

89. R. H. Holm, Chern. Rev., 1987,87, 1401.

90. K. D. Karlin and Y. Gultneh, Prog. Inorg. Chern., 1087, 35, 219.

91. K. O. Karlin, M. S. Haka, R. W. Cruse and Y. Gultneh, J. Am. Chern. Soc.,

1985,107,5828. J

92. K. D. Karlin, M. S. Haka, R. W. Cruse, G. J. Meyer, A. Farooq, Y. Gultneh,

J. C. Hayes and J. Zubieta, J. Am. Chern. Soc., 1988, 110, 1196.

93. N. J. Blackburn, R. W. Strange, A. Faraoq, M. S. Haka and K. O. Karlin, J.

Am. Chern. Soc., 1985,110,4263.

94. K. O. Karlin, Z. Tyeklar, A. Faraoq, R. R. Jacobson. E. Sinn, D. W. Lee, J.

E. Bradshaw and L. J. Wilson, fnorg. Chem. Acta, 1991, 182, 1.

95. F. Zippel, F. Ahlers, R. Werner, W. Haase, H-F. Nolting and B. Krebs,

Inorg. Chem., 1996, 35, 3409.

96. E. Muller, G. Bernardinelli and J. Reedjik, Inorg. Chern., 1995,34,5979.

97. N. Kitajima and Y. Mora-oka, Chern. Rev., 1994, 94, 737.

98. Y. Nishida, H. Shimo, J. Maehara and S. Kida, J. Chem. Soc., Dalton

- .

211

Trans., 1985, 1945

99. S. K. Mandai, L. K. Thompson, M. J. Newlands, A. K. Biswas, B. Adhikary,

K. Nag, E. J. Gabe and F. L. Lee, Can. J. Chem.,1989, 67, 662.

100. S. S. Tandon, L. K. Thompson, J. N. Bridgson, V. McKee and A. J.

Downward, Inorg. Chem., 1992,31,4635.

101. S. M. Nelson, F. Esho, A. Lavery and M. G. B. Drew, J. Am. Chem. Soc.,

1983, 105,5693.

102. 8. L. Feringa in Bioinorganic Chemistry of Copper, eds K. D. Karlin and Z.

Tyeklar, Chapman and Hall, New York, 1993, p.306.

103. M. Reg/ier, C. Jorand and B. Waegell, J. Chem. Soc., Chem. Commun.,

1990,1752.

104. L. Casel/a, M. Gullotti, R. Radael/i, P. Di Gennaro, J. Am. Chem. Soc.,

Chern. Commun., 1991, 1611.

105. J. Reim and B. Krebs, J. Chem. Soc., Dalton Trans., 1997,3793.

106. L. Casella, M. Gullotti, M. Bartosek, G. Pallanza and E. Laurenti, J. Chem.

Soc., Chem. Commun., 1991, 1235.

107. L. Casella and M. Gullotti in Bioinorganic Chemistry of Copper, eds K. D.

Karlin and Z. Tyeklar, Chapman and Hall, New York, 1993, p.292.

108. G. Alzuet, L. Casella, M. L. ViI/a, O. Carugo and M. Gul/otti, J.. Chem. Soc.,

Dalton Trans., 1997,4789. .I

109. D. E. Fenton in Advances in Inorganic and Bioinorganic Mechanisms, Vol.

2, ed. A. G. Sykes, Academic press, London, 1983, p.187.

110. K. D. Karlin and J. Zubieta, Biological and Inorganic Copper Chemistry,

Vols. 1 and 2, Adenine Press, 1983.

111. D. E. Fenton, U. Casel/ato, P. A. Vigato and M. Vidali, Inorg. Chim. Acta,

1984,95,187.

112. D. E. Fenton, U. Casel/ato, P. A. Vigato and M. Vidali, Inorg. Chim. Acta,

1982,6257.

-113. U. Casellato, P. A. Vigato, D. E. Fenton and M. Vidali, Chem. Soc. Rev.,

1979, 8, 199.

114. S. E. Groh, Isr. J. Chem., 1976/1977, 15,277.

- .

115. D. M. Dooley, R. A. Scott, E. Ellinghaus, E. I. Solomon and H. B. Gray,

Proc. Nat/. A cad. Sci. U. S. A., 1978,75,3019.

116. D. J. Spira, M. E. Winkler and E. I. Solomon, Biochem. Biophys. Res.

Commun., 1982, 107, 721.

212

117. C. D. LuBien, M. E. Winkler, T. J. Thamann, R. A. Scott, M. S. Co, K. O.

Hodgson and E. I. Solomon, J. Am. Chem. Soc., 1981, 103,7014.

118. J. A. Fee and R. G. Briggs, Biochim. Biophys. Acta, 1975, 400, 439.

119. J. A. Fee, J. BioI. Chem., 1973,248,4229.

120. L. Calabrese, D. Cocco and A. Desideri, FEBS Lett., 1979, 106, 142.

121. E. L. Mutterties, T. N. Rhodin, E. Band, G. F. Brucker and W. R. Pretzer,

Chem. Rev., 1979, 79, 91.

122. E. C. Niederhoffer, J. H. Timmons and A. G. Martell, Chem. Rev., 1984,

84, 137.

123. P. Cadiot and W. Chodkiewicz in Chemistry of Acetylenes, ed. H. G.

Viehe, Marcel Dekker, New York, 1969, p.597.

124. A. S. Hay, J. Org. Chem., 1962,27,3320.

125. H. L. Finkbeiner, A. S. Hay and D. M. White in Polymerisation by Oxidative

Coupling, eds. C. E. Schildnecht and I. Skeist, Interscience, New York,

1977, p. 537-581.

126. C. Glaser, Ann. Chem. Pharm., 1870,154,159.

127. W. G. Nigh in Oxidation in Organ;c Chemistry, Part B, eds. W. S.

Trachanovsky, Academic, New York, 1973, Pp.1-96.

128. M. M. Rogic and T. R. Demmin, J. Am. Chem. Soc., 1978, 100,5472.

129. W. W. Kaeding, Hydrocarbon Proc., 1964,43, 173.

130. W. Brenner, U. S. Pat. 3, 1974, 796,732.

131. E. L. Reilly, U. S. Pat. 3,1976,987,068.

132. K. D. Karlin, Z. Tyeklar, A. Farooq, M. S. Haka, P. Ghosh, R.W. Cruse,

Y. Gultneh, J. C. Hayes, P. J. Toscano and J. Zubieta, Inorg.Chem.,

1992,31,1436.

133. W. P. J. Gaykema, W G. J. Hoi, J. M. Vereijken, N. M. Soeter, H. J. Bak

and J. J. Beintema, Nature, 1984,309,23.

134. K. Lerch, Life Chern. Reports, 1987,5,221.

135. L. Casella and L. Rigoni, J. Chern. Soc. Commun., 1985, 1668

136. L. Casella, M. Gullotti, G. Pallanza and L. Rigoni, J. Am. Chern. Soc.,

1988,110,4221.

137. K. D. Karlin and J. K. Yandell, Inorg. Chern., 1984,23,1184.

138. S. G. Burton, Ph.D Thesis, Rhodes University, 1993.

139. H. M. J. Hendriks, P. J. M. W. L. Birker, J. van Rijn, G. C. Verschoor

and J. Reedjik, J. Am. Chern. Soc.,1982, 104,3607.

140. P. S. Bailey and R. E. Erickson, Org. Synth., 1961,41,41.

213

141. B. S. Furniss, A J. Hannaford, P. W. G. Smith and A R. Tatchell, Vogel's

Textbook of Practical Organic Chemistry, 5th ed. 1989, Longman Scientific

and Technical, p.1061

142. M. L. Bode, P. T. Kaye and R. George, J. Chern. Soc., Perkin Trans.,

1994,3023.

143. T. N. Sorrell and M. L. Garrity, Inorg. Chern., 1991,30,210.

144. G. F. Smith and W. H. McCurdy Jr., Anal. Chern., 1952,24,371.

145. C. J. Chandler, L W. Deady and J. A Reiss, J. Heterocyciic Chern., 1981,

18,599.

146. A. D. Baylis and M. E. D. Hillman, German Patent 2155113 (1'972).

147. S.E. Drewes and N. D. Emslie, J. Am. Chern. Soc.,Perkin Trans. I, 1002,

2079.

148. M. L Bode and P. T. Kaye, J. Chern. Soc., Perkin Trans., 1993, 1809.

149. O. B. Familoni, P. T. Kaye and P. J. Klaas, Chern. Commun., 1998,,2563.

150. R. S. Robinson, Ph.D Thesis, Rhodes University, 1997.

151. M. G. B. Drew, C. P. Waters, S. G. McFall and S. M. Nelson, J. Chern.

Res., 1979,360.

152. A. Bencini, A Bianchi, E. Garcia-Espana, M. Micheloni and P. Paoletti,

Inorg. Chern., 1988, 27, 176.

_. 153. A. Bencini, A Bianchi, E. Garcia-Espana, S. Mangani, M. Micheloni, P.

Orioli and P. Paoletti,1norg. Chern., 1988,27, 1104.

154. R. Menif and A E. Martell, J. Chern. Soc., Chern. Commun.,1989, 1521.

214

155. F. Cabral, B. Murphy and J. Nelson, J. Inorg. Acta., 1984,90, 169.

156. J. M. Lehn, Angew. Chem., 1988, 100,91; Angew. Chem., Int. Ed. Engl.,

1988,27,90.

157. H. Jahansouz, Z. Jiang, R. H. Himes, M. P. Hertes and K. B. Mertes, J.

Am. Chem. Soc., 1989, 111, 1409.

158. N. Bergman and F. P. Schmidtchen, Tetrahedron Lett., 1988,29,6235.

159. E. Kimura, Y. Kuramoto, T. Koike, H. Fijioka and M. Kodama, J. Org.

Chern., 1990, 55, 42.

160. K. E. Krakowiak, J. S. Bradshaw, W. Jiang, N. K. Dalley, G. Wu and R. M.

Izatt, J. Org. Chern., 1991, 56, 2675.

161. K. E. Krakwiak, J. S. Bradshaw, W. Jiang, N. K. Dalley, G. Wu and R. M.

Izatt, J. Org. Chem., 1991, 56, 2675.

162. C. Kallner, B. Pugin and A. Togni, J. Arn. Chern. Soc., 1998, 120,10274.

163. D. A. Tomalia, A. M. Taylor and W. A. Goddard, Angew. Chern., Int. Ed.

Engl., 1990,29,138.

164. D. A. Tomalia, A. M. Taylor and W. A. Goddard, Angew. Chern., 1990,

102,119.

165. G. R. Newkome and C. N. Moorefield, Aldrichchirn. Acta.,1992, 25, 31.

166. R. Engel, Polyrner News, 1992, 17,301. ,I

167. H. G. Mekelburger, W. Jaworek and F. Vagtle, Angew. Chem., 1992,

104,1609.

168. D. A. Tomalia, Adv. Mater., 1994,6,530.

169. P. Lange, A. Schier and H. Schmidbaur, Inorg. Chern., 1996,35,637.

170. R. Dagani, Chernical and Engineering News, 1996, June 3, p. 30.

171. G. R. Newkome, C. N. Moorfield and K. J. Theriot, J. Am. Chern. Soc.,

1988, 53, 5552.

172. X. A. Dominguez, I..G. Lopez and R. Franco, Notes, 1961, 1625.

_. _ 173. P. N. W. Baxter, J. A. Connor, J. D. Wallis, D. C.-Povey and A. K. Powell,

J. Chern. Soc., Perkin Trans, 1, 1992, 1601. ~ .- ;;

174. W. Paw and R. Eisenberg, Inorg. Chern, 1997, 36, 2287.

175. C. Hiort, P. Lincoln and B. Norden, J. Am. Chern. Soc., 1993, 115,3448.

215

176. Y. Singh, R Sharan and RN. Kapoor, Indian Journal of Chemistry, Sect.

A, 1986, 25A, 771.

177. S. C. Dixit, R Sharan, and RN. Kapoor, J. Organomet. Chem., 1987,

332, 135.

178. A. K. Narula and P. Lukose, J. Organomet. Chem., 1990, 393, 365.

179. S. P. Datta and B. R Rabin, J. Chem. Soc., Faraday Trans., 1956, 52,

1123.

180. S. K. Kim and A. E. Martell, J. Am. Chem. Soc., 1966,88,914.

181. R B. Martin, M. Chamberlain and J. E. Edsall, J. Am. Chem. Soc., 1960,

82, 495

182. E. J. Billoand D. W. Margerum, J. Am. Chem. Soc., 1970,92,6811.

183. H. C. Freeman and M. L. Golomb, J. Chem. Soc., Chem. Commun.,

1970,1523.

184. E. B. Wilson and R B. Martin, Inorg. Chem., 1970,9,528.

185. J. C. Cooper, L. F. Wong and D. W. Margerum, Inorg. Chem., 1978,17,

261.

186. D.A. Phillips, Metals and Metabolism, Clarendon, Oxford, 1976.

187. R Williams, The Metals of Life, Van Nostrand Reinhold, New York, 1971.

188. A. Massey in Comprehensive Inorganic Chemistry, ed. J. C. eailar, H. J.

Emeleus, R S. Nyholm and A. F. Troman-Dickinson, Pergamon, Oxford,

1973, vol. 3, p. 1.

189. N. V. Sidgwick in Comprehensive Coordination Chemistry, ed. G.

Wilkinson, RD. Gillard and J. A. McCleverty, Pergamon, Oxford, 1987,

Vol. 5, p. 535.

190. A. F. Wells in Comprehensive Coordination Chemistry, ed. G. Wilkinson,

RD. Gillard and J. A. McCleverty, Pergamon, Oxford, 1987, vol. 5, P 535.

191. J. D. Dunitz and L. E. Orgel, J. of Phys. Chem. Solids, 1957, 3, 20.

192. H. A. Jahn and E. Teller, Proc. R. Soc. London, 1937, 161,220.

193. F. H. Jardin, Adv.lnorg. Chem. Radiochem., 1975, 17,115.

194. W. E. Hatfield and R Whyman, Transition Met. Chem., 1969, 47, 5.

195. F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, Wiley

Interscience, New York, 5th ed., 1980, Pp757,768.

196. A. D. Cross and R. A. Jones, An Introduction to Practical Infrared

Spectroscopy, 3rd ed. London Butterworths, 1969, p.97.

197. K. Nakamoto and S. J. McCarthy, Spectroscopy and Structure of Metal

Chelate Compounds, John Wiley and Sons, 1968, p. 265.

198. J. D. Lee, Concise Inorganic Chemistry, 4th ed., Chapman and

Hall, London, 1991, Pp. 783.

199. Reference 203 Pp. 788.

200. Reference 203 Pp. 792.

201. Reference 203 Pp. 796.

202. M. N. Hughes, The Inorganic Chemistry of Biological Processes, John

Wiley and Sons, London, 1975, p.186.

203. D. E. Fenton, Biocoordination Chemistry, Oxford University Press, New

York, 1995, Pp. 69, 70.

204. M. N. Hughes, The Inorganic Chemistry of Biological Processes, John

Wiley and Sons, London,1975, Pp. 243, 245-250.

205. D. E. Fenton, Biocoordination Chemistry, Oxford University Press, New

York, 1995, p. 21.

206. G. Mclendon and A. E. Martell, Coord. Chem. Rev., 1976, 19~ 1.

207. R. W. Hay, Coord. Chem. Rev., 1981,35,85. !

216

208. A. Bakac, B. Assink, J. H. Espol1Son and W. D. Wang, Inorg. Chem., 1996,

35,788.

209. J. A. Lee-Thorp, J. E. ROede and D. A. Thornton, 1978,50,65.

210. D. Nicholls, Complexes and First Row Transition Elements, The MacMillan

Press, London, 1974, Pp. 96.

211. C. Pariya, S. Ghosh, A. Ghosh, M. Mukherjee, A. K. Mukherjee and N. R.

Chaudhuri, J. Chem. Soc. Dalton Trans., 1995, 337

212. F. A. Cotton, G. WilRinson and P. L. Gaus, Basic Inorganic Chemistry, 2nd

ed. 1987, John Wiley and Sons, New York, p.502-504.

213. D. Volkmer, A. Horstman, K. Griesar, W. Haase and B. Krebs, Inorg.

Chem., 1996, 35,1132.

217

214. Reference 78, Pp.78-80.

215. Reference 194, Pp.186

216. Y. Ihara, Y. Fukuda and K. Sone, Inorg. Chem., 1987,26,3745.

217. Reference 215, Pp.91.

218. Reference 215, Pp.95.

219. Reference 195, Pp.353.

220. Reference 217, Pp.537.

221. N. Farrell, Transition Metal Complexes as Drugs and Chemotherapeutic

Agents in Catalysis by Metal Complexes, eds B. R. James and R. Ugo,

Kluwer Academic Publishers, Dordrecht , 1989, Vol. 11, Pp.46-66

222. C. A. Hunter and J. K. M. Sanders, J. Am. Chem. Soc., 1990, 112, 5525.

223. M. Van Beusichem and N. Farrell, Inorg. Chem., 1992,31,634.

224. N. L. Alinger, Adv. Phys. Org. Chem. ;"1976, 13.1.

225. N. L. Alinger, J. Am. Chem. Soc., 1977,99,8127.

226. D. J. Duchamp, A. Chem. Soc. Symp. Ser., 1979, 112,79; M. Geller and

D. Shugar, Drugs Exp. Clin. Res. 1986,12,595.

227. R. F. Tilton, P. K. Weiner and P. A. Kollman, Biopolymers, 1983, 22, 969.

228. M. R. Snow, J. Am. Chem. Soc., 1970,92,3610.

229. R. B. Hancock, Prog.lnorg. Chem., 1989,37,187.

230. J. C. A. Boeyens, C. C. Fox and R. B. Hancock, Inorg. Chim. Acta., .1

1984,87,1.

231. P. V. Bernhardt and P. Comba, Helv. Chim. Acta., 1991, 74, 1834; 1992,

75,645.

232. M. Yamaguchi, S. Yamamatsu, T. Furusawa, S. Yano, M. Saburi and S.

Yoshikawa, Inorg. Chem., 1980,19,2010.

233. R. B. Hancock, P. W. Ngenya, A. Evers, P. W. Wadem and J. C. A.

Boeyens, Inorg. Ch~m., 1990, 29, 264.

234. N. Oishi, Y. Nishida, K. Ida and S. Kida, Bull. Chem. Soc. Japan, 1980, 53,

2847.

235. E. I. Solomon, International symposium on oxidases and related redox

systems, 1988, 309.

218

236. N. Kitajima and Y. Moro-aka, J. Chem. Soc. Dalton Trans., 1993,2665.

237. A. Berces, Inorg. Chem., 1997,36,4831.

238. S. Kida, H. Okawa and Y. Nishida in Copper Coordination Chemistry:

Biochemical and Inorganic perspectives, eds K. D. Karlin and Y. Zubieta,

Adenine, Guilderland, New York, 1983, p.425.

239. D. E. Wilcox, A. G. Porras, Y. T. Hwang, K. Lerch, M. E. Winkler and E. I.

Solomon, J. Am. Chem. Soc., 1985, 107,4015.

240. B. Jung, K. D. Karlin and A. D. Zuberbuhler, J. Am. Chem. Soc., 1996,

118,3763.

241. D-H Lee, N. Wei, N. N. Murthy, Z. Tyeklar, K. D. Karlin, S. Kaderli, B.

Jung and A. D. Zuberbuhler, J. Am. Chem. Soc., 1995, 117, 12498.

242. Z. Tyeklar, R. R. Jacobson, N. Wei, N. N. Murthy, J. Zubieta and K. D. ',,",

Karlin, J. Am. Chem. Soc., 1993, 115,2677.

243. M. Reglier in , Bioinorganic Chemistry of Copper, eds. K. D. Karlin and Z.

Tyeklar, Chapman and Hall, New York, 1993, p.348.

244. B. W. Rossiter and J. F. Hamilton, Physical methods of chemistry, 2nd ed.

1986, Vol. 2, Electrochemical methods, Wiley, New York.

245. P. Zanello, S. Tamburini, A. Vigato and G. A. Mazzocchin, Coord. Chem. , Rev., 1987, 77,165.

,I

246. D. E. Fenton, R. R. Schroeder and R. L. Lintveldt, J. Am. Chem. Soc.,

1978,100, (6),1931.

247. R. R. Gagne, R. P. Kreh and J. A. Dodge, J. Am. Chem. Soc., 1979,

101(23),6917.

248. G. Christou, S. P. Perlepes, E. Libby, K. Folting, J. C. Huffmann, R. J.

Webb and D. N. Hendrickson, Inorg. Chem., 1990,29,3657.

249. Reference 68, Pp. 371.

250. M. Ali, I. Zilberman,...H. Cohen, A. I. Shames and D. Meyerstein, Inorg.

Chem., 1996,35,5127.

251. K. J. Oberhausen, _R,;. J. 0' Brien, J. F.< Richardson and R. M. Buchanan,

Inorg. Chim. Acta., 1990, 173, 145.

252. R. G. Pearson, Science, 1966, 151, 172.

219

253. J. P. Chyn and F. L. Urbach, Inorg. Chim. Acta., 1991,189,157.

254. K. T. Yasunobu, E. W. Peterson and H. S. Mason, J. Bioi. Chem., 1959,

234, 3291.

255. Passi and M. Nazarro-Porro, Brit. J. Oermatol., 1981, 104,659.

256. P. Paul, Z. Tyeklar, R. Jacobson and K. D. Karlin, J. Am. Chem. Soc.,

1991,113,5322.

257. K. Kushioka, J. Org. Chem., 1984,49,4456.

258. M. Flinterman and G. Challa, J. Mol. Cata/., 1983, 18, 149.

259. S. Tsuruya, Y. Kishikawa, R. Tanaka and T. Kuse, J. Catal., 1977,49,254.

260. B. Feringa and H. Wynberg, Bioorg. Chem., 1978,7,397.

220

5 APPENDIX

Crystallographic data for cobalt complex 136.

Table 1: Crystal data and structure refinement for cobalt complex 136.

Identification code Empirical formula Formula weight Temperature Wavelength Crystal system Space group Unit cell dimensions

Volume Z Density (calculated) Absorption coefficient F(OOO) Theta range for data collection Limiting indices Reflections collected 1 unique Independent reflections Completeness to theta Absorption correction Refinement method Data 1 restraints / parameters Goodness-of-fit on F2 Final R indices [1>2sigma (I)] R indices (all data) Largest diff. peak and hole

rw6rco m C21 H2o CI2 Co N2 430.24 296(2) K 0.71073 A Monoclinic P(2)/n a = 8.7818 (5) A alpha = 90 deg b = 16.0994 (10) A beta = 91.8600 {"10) deg. c = 13.7265 (8) A gamma = 90 deg. 1939.7 (2) A3

1.892 Mg/m3

1.193 mm-1

1148 1.95 to 28.30 deg. -9<=h<=11, -21<=k<=19, -17<=1<=17 11620 4349 [R (int) = 0.0386] , 28.30, 90.1 % None Full-matrix least-squares on F2 4349/0/315 1.074 R1 = 0.0581, wR2 = 0.1477 .. R1 = 0.0992, wR2 = 0.1730 0.315 and -0.302 e.A3

221

Table 2: Atomic coordinates (x 104) and equivalent isotropic dislacement

parameters (A2 x 103) for cobalt complex 136.a

,b U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

x y z U (eq)

Co(l) 2209(1) 1070(1) 2078(1) 46(1) Cl (1) 2963(2) 2060(1) 3119 (1) 83(1) N (1) 4200(3) 583(2) 1632(2) 45 (1) C (1) 5297(5) 1115(3) 1380(3) 52 (1) Cl(2) 833(1) 1373(1) 715 (1) 65(1) N (2) 1283(3) 41(2) 2754(2) 37 (1) C(2) 6622(5) 850(4) 963 (3) 57(1) C (3) 6815(5) 19(3) 796(3) 59(1) C(4) 5724(5) -526(3) 1092(3) 52 (1) C(5) 4408(4) -238(2) 1508(3) 44 (1) C (6) 3153(6) -815(3) 1780(4) 54(1) C (7) 2420(5) -651(3) 2758(3) 48(1) C (8) -154(4) -212c(3) 2214(3) 41 (1) C (9) -1205(4) -794(2) 2725(3) 39 (1) C(10) -1405(5) -1611(3) 2395(3) 48 (1) C(ll) -2489(5) -2118(3) 2793(4) 58 (1) C(12) -3398(5) -1820(3) 3510(3) 59 (1) C(13) -3216(5) -1015(3) 3849(3) 52 (1) C(14) -2131(4) -491(2) 3460 (3) 41(1) C(15) -1936(4) 377(2) 3822(3) 43(1) C(16) -3205(5) 840(3) 4070(3) 56(1) C(17) -3038(6) 1627(3) 4470(3) 62 (1) C (18) -1627(6) 1949(3) 4620(4) 64 (1) C(19) -349(6) 1520(3) 4360(3)~ 54(1) C(20) -480(4) 719(2) 3960(3) 43 (1) C (21) 933(4) 209(3) 3800(3) .I 42 (1)

a For atom labelling see Figure 36.

b Estimated standard deviations in parenthesis.

222

Table 3: Bond lengths [AJ and angles [deg] for cobalt complex 136.

Co (1) -N (1) 2.030(3) Co(1)-N(2) 2.078(3) Co (1) -Cl (1) . 2.2261(14) Co(1)-Cl(2) 2.2481(12) N(l) -C(l) 1.342(5) N(l)-C(5) 1.345(5) C(l)-C(2) 1.381(6) N(2) -C(7) 1.496(5) N(2) -C (8) 1.499(5) N (2) -C (21) 1.502(5) C(2)-C(3) 1.370(7) C(3)-C(4) 1.370(6) C(4)-C(5) 1.386(5) C (5) -C (6) 1.499(6) C(6) -C(7) 1.530(6) C(8) -C(9) 1.505(5) C(9)-C(10) 1.400(5) C(9)-C(14) 1.404(5) C(10)-C(11) ~, 1.379 (6) C(11)-C(12) 1.374 (7) C(12)-C(13) 1.385(6) C(13)-C(14) 1.392 (5) C(14)-C(15) 1.491(5) C(15)-C(16) 1.391(6) C(15)-C(20) 1.399(5) C(16)-C(17) 1.387(6) C(17)-C(18) 1.353(7) C(18)-C(19) 1.375 (7) C(19)-C(20) 1.404(6) C(20)-C(21) 1.510(5)

N(1)-Co(1)-N(2) 100.36(12) .I

N (1) - Co ( 1) - Cl (1) 103.21(10) N(2)-Co(1)-Cl(1) 113.38(9) N(1)-Co(1)-Cl(2) 106.14(9) N(2)-Co(1)-Cl(2) 109.74(9) Cl(1)-Co(1)-Cl(2) 121.30(6) C(1)-N(l)-C(5) 119.5(3) C(l)-N(l)-Co(l) 117.6(3) C (5) - N (1) - Co (1) 122.6(3) N(l) -e(l) -C(2) 122.1(5) C(7) -N(2) -C(8) 110.5(3) C(7) -N(2) -C(21) 106.8(3) C(8) -N(2) -C(21) 109.1(3) C (7) - N (2) - Co (1) 109.0(2) C(8)-N(2)-Co(1) 109.3(2)

- . C(21)-N(2)-Co(1) 112.2(21 -C(3) -C(2) -C(l) 118.7(5) C(2) -C(3) -C(4) 119.0(4) C(3) -C(4) -C(5) c

120.5(4) N(l) -C(5) -C(4) 119.9(4) N(l) -C(5) -C(6) 118.3'( 3)

C(4) -C(5) -C(6) C(5) -C(6) -C(7) N(2) -C(7) -C(6) N(2) -C(8) -C(9) C(10) -C(9) -C(14) C(10) -C(9) -C(8) C(14) -C(9) -C(8) C(ll) -C(10) -C(9) C(12)-C(11)-C(10) C(11)-C(12)-C(13) C(12)-C(13)-C(14) C(13)-C(14)-C(9) C(13) -C(14) -C(15) C(9) -C(14) -C(15) C(16)-C(15)-C(20) C(16)-C(15)-C(14) C(20)-C(15)-C(14) C (17) -C (16) -C (15) C(18)-C(17)-C(16) C(17)-C(18)-C(19) C(18)-C(19)-C(20) C(15)-C(20)-C(19) C(15)-C(20)-C(21) C(19)-C(20)-C(21) N(2) -C(21) -C(20)

121. 6 (4) 116.4(4) 115.1(3) 117.2(3) 119.3(4) 120.5(3) 119.8(3) 120.6(4) 120.1(4) 120.3(4) 120.7(4) 119.0.( 4) 120.8(3) 120.2(3) 119.5(4) 119.9(4) 120.5(3) 120.7(5) 119.7(5) 121.3(5) 120.4(4) 118.5(4) 121.2 (3) 119.9(4) 115.6(3)

223

. Co (1) Cl(l) N (1) C(l) Cl(2) N (2) C(2) C (3) C(4) C(5} C (6) C (7) C(8) C(9) C(10) C(ll) C(12) C(13)

. C (14) C(15) C(16) C(17) C(18) C(19) C (20) C (21)

Table 4: Anisotropic displacement parameters (A2 x 103) for cobalt complex

136. The anisotropic displacement factor exponent takes the form: -2 pi2 [h2 a*2 U11 + ... + 2 h k a* b* U12]

Ull U22 U33 U23 U13

40(1) 34(1) 64(1) 6 (1) 15 (1) 76(1) 53 (1) 122 (1) -30(1) 39 (1) 36(2) 44(2) 54(2) -4(2) 7 (1) 42(2) 59(3) 56(3) - 5 (2) 4 (2) 56 (1) 66 (1) 74 (1) 30(1) 16 (1) 35(2) 36(2) 39(2) 4 (1) 3 (1) 37(2) 84(4) 50 (3) -5 (2) 3 (2) 37(2) 89(4) 50(2) - 8 (2) 1 (2) 48(2) 60(3) 47(2) -4(2) - 6 (2) 42(2} 45(2) 45(2) -2(2) o (2) 56(3) 38(2) 69(3) - 6 (2) 14(2) 50(2) 40(2) 53(2) 14(2) 6 (2) 40(2) 46(2) 36(2) .. ~ 0(2 ) -1 (2) 39(2) 42(2) 36(2) 5 (2) - 3 (2) 52(2) 45(2) 48(2) -1(2) - 3 (2) 65(3) 40(2) 69(3) 2 (2) -7 (2) 58(3) 55(3) 65(3) 12(2) 6 (2) 52(2) 58(3) 48 (2) 4(2) 10(2) 42(2) 44(2) 36(2) 7 (2) 4 (2) 47(2) 45(2) 38 (2) 5 (2) 5 (2) 50(3) 65(3) 54(3) -2(2) 11(2) 75(3) 57 (3) 55(3) -4(2) 22( 2) 91(4) 44(3) 59(3) - 6 (2) 21(3) 70(3) 49(2) 44(2) - 3 (2) 1,8(2) 53(2) 46(2) 30(2) 2 (2) 7 (2) 45(2} 43 (2) 39 (2) 7 (2) -2 (2) .I

224

U12

3 (1) -20 (1)

o (1) -7 (2) 18(1)

1 (1) - 8 (2) 12(2) 19(2} 10(2}

9 (2) 6 (2) 1 (2)

-4 (2) -1 (2)

-10(2) -19(2)

-7 (2) - 5 (2) -5 (2) 1(2)

11 (3) -4 (3)

-13 (2) - 3 (2) -9 (2)

225

Table 5: Hydrogen coordinates ( x 104) and isotropic displacement parameters

(A2 x 103) for cobalt complex 136.

x y z U(eq)

H (1) 5160(50) 1690(30) 1470(30) 63 (13) H(2) 7310(60) 1220(30) 850(40) 79(17) H (3) 7650(60) -220(30) 480(40) 76(15) H (4) 5830(50) -1100(30) 1030(30) 60(13) H(6B) 2310(50) -790(20) 1240(30) 50(11) H (6A) 3510(60) -1270(30) 1780(30) 73(16) H(7B) 2000(50) -1150(30) 3050(30) 48(11) H (7A) 3160(50) -490 (30), 3210(30) 48(11) H(8B) -720(40) 230(20) 2100(30) 37(10) H (8A) 130(40) -470(20) 1530(30) 40 (9) H (10) -860(50) -1790(30) 1900(30) 51(12) H(ll) -2600(50) -2640(30) 2520(30) 57(12) H(12) -4140(60) -2150(30) 3780(40) 87(17) H(13) -3790(50) -790(30) 4390(30) 56(12) H(16) -4190(50) 650(30) 3920(30) 49(11) H(17) -3800(50) 1910(30) 4640(30) 53 (12) H (18) -1480(60) 2420(40) 4850(40) 82(17) H(19) 750(50) 1720(30) 4480(30) 54(12) H{21B) 830(30) -320(20) 4096(20) 22(8) H (21A) 1820(40) 480(20) 4080(30) 41(10)

J

226

Crystallographic data for nickel complex 143.

Table 1: Crystal data and structure refinement for nickel complex 143.

Identification code Empirical formula Formula weight Temperature Wavelength Crystal system Space group Unit cell dimensions

Volume Z Density (calculated) Absorption coefficient F(OOO) Crystal size Theta range for data collection Limiting indices Reflections collected / unique Independent reflections Refinement method Data / restraints / parameters Goodness-of-fit on F2 Final R indices [1>2sigma (I)] R indices (all data) Largest diff. peak and hole

newfin C21 H2o CI2 Co N2 430.00 296(2) K 0.71073 A Monoclinic P(2)/n a = 8.7689 (5) A alpha = 90 deg b = 15.8377 (10) A beta = 91.8760 (10)deg. C = 13.7979 (8) A gamma = 90 deg. 1915.2 (2) A3

1'.491 Mg/m3

1.299 mm-1

888 0.20 x 0.20 x 0.10 mm 1.96 to 28.28 deg. --11=h<=10, -20<=k<=20, -17<=1<=16 11748 4317 [R (int) = 0.0328] Full-matrix least-squares on F2 4317 / 0 / 31 5 ~ 1.087 J

R1 = 0.0494, wR2 = 0.0979 R1 = 0.0714, wR2 = 0.1072 0.367 and -0.339 e.A3

227

Table 2: Atomic coordinates ( x 104) and equivalent isotropic dislacement

parameters (A2 x 103) for nickel complex 143.a

,b U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

x y z U(eq)

Ni (1) 2197(1) 1086 (1) 2108(1) 41 (1) Cl(l) 3094(1) 1985(1) 3225(1) 68 (1) N (1) 4142(3) 589(2) 1657(2) 39 (1) C (1) 5243(4) 1134(2) 1417(3) 46 (1) Cl(2) 861(1) 1315(1) 718(1) 56 (1) C(2) 6558(4) 879(3) 991(3) 51 (1) N (2) 1270(3) 46(2) 2744(2) 33 (1) C(3) 6770(4) 38(3) 816(2) 50 (1) C(4) 5679(4) -526(2) 1082(2) 46 (1) C(5) 4344(4) -239(2) 1499(2) 40 (1) C (6) 3092(4) -831(2) 1754(3) 48 (1) C(7) 2390(4) -671(2) 2734(2) 42 (1) C (8) -192(3) -18~(2) 2207(2) 35 (1) C(9) -1226(3) -791(2) 2711(2) 34 (1) C(10) -1432(4) -1611(2) 2366(2) 42 (1) C (11) -2527(4) -2130(2) 2755(3) 51 (1) C(12) -3421(4) -1840(2) 3487(3) 53 (1) C(13) -3223(4) -1030(2) 3839(3) 47(1) C(14) -2125(3) -493(2) 3459(2) 36 (1) C(15) -1930(3) 380(2) 3839(2) 37 (1) C (16) -3192(4) 858(2) 4095(3) 48 (1) C(17) -3015(5) 1644(2) 4514(3) 55 (1) C(18) -1582(5) 1973(2) 4668(3) 55 (1) C(19) -317(4) 1528(2) 4391(2), 47(1) C (20) -465(4) 728(2) 3968(2) 38 (1) C (21) 932(4) 202(2) 3792(2) .I 37 (1)

a For atom labelling see Figure 46. b Estimated standard deviations in parenthesis.

228

Table 3: Bond lengths [A] and angles [deg] for nickel complex 143.

Ni(l)-N(l) 1.997(2) Ni(1)-N(2) 2.047(2) Ni(l)-Cl(l) 2.2229(11) Ni(1)-Cl(2) 2.2447(10) N(l) -C(5) 1.342(4) N (1) -C (1) 1.344(4) C(1)-C(2) 1.372 (5) C(2)-C(3) 1.368(5) N(2)-C(7) 1.501(4) N(2)-C(8) 1.505(4) N(2) -C(21) 1.506(4) C(3)-C(4) 1.368(5) C(4)-C(5) 1.397(4) C(5)-C(6) 1.495 (5) C(6)-C(7) 1.525 (5) C(8)-C(9) 1.506 (4) C(9)-C(10) 1.393(4) C(9)-C(14) 1.401(4) C(10)-C(11) 1.386(5) C(11)-C(12) 1.377 (5) C(12)-C(13) 1.381(5) C(13)-C(14) 1.399(4) C (14) -C (15) 1.487 (4) C(15)-C(16) 1. 395 (4) C(15)-C(20) 1.403(4) C(16)-C(17) 1.380(5) C(17)-C(18) 1.3~0(6) C(18)-C(19) 1.379(5) C(19)-C(20) 1.400(5) C(20)-C(21) 1.508(4)

,I N(1)-Ni(1)-N(2) 99.87(10) N(l)-Ni(l)-Cl(l) 100.49(8) N(2)-Ni(1)-Cl(1) 110.81(7) N(1)-Ni(1)-Cl(2) 102.97(8) N(2)-Ni(1)-Cl(2) 107.03(7) Cl(1)-Ni(1)-Cl(2) 130.68(4) C(5) -N(l) -C(l) 119.2(3) C(5)-N(1)-Ni(1) 123.7(2) C(l) -N(l) -Ni (1) 116.8(2) N(1)-C(1)-C(2) 122.5(3) C(3)-C(2)-C(1) 118.9(4) C(7) -N(2) -C(8) 111.1(2) C(7) -N(2) -C(21) 106.4(2) C(8) -N(2) -C(21)-- 108.8'( 2) C(7)-N(2)-Ni(1) 109.6(2)

- C(8) -N(2) -Ni (1) 109.2(2-) -C(21) -N(2) -Ni (1) 111.8 (2) C(4) -C(3) -C(2) ~119.2(3)

C(3) -C(4) -C(5) 120.0(3) N(l) -C(5) -C(4) 120.1(3) N(l) -C(5) -C(6) 118.3(3)

C(4)-C(S)-C(6) C(S) -C(6) -C(7) N(2) -C(7) -C(6) N(2)-C(8)-C(9) C(10)-C(9)-C(14) C(10)-C(9)-C(8) C(14)-C(9)-C(8) C(ll) -C(10) -C(9) C(12) -C(ll) -C(10) C(ll) -C(12) -C(13) C(12) -C(13) -C(14) C(13) -C(14) -C(9) C(13)-C(14)-C(lS) C(9) -C(14) -C(1S) C (16) -C (IS) -C (20) C (16) -C (1S) -C (14) C(20)-C(lS)-C(14) C(17) -C(16) -C(lS) C(18) -C(17) -C(16) C(17)-C(18)-C(19) C(18) -C(19) -C(20) C(19) -C(20) -C(lS) C (19) -C (20) -C (21) C(lS)-C(20)-C(21) N(2) -C(21) -C(20)

121.6(3) 11S.1(3) 114.6(3) 116.3(2) 119.7(3) 120.6(3) 119.2(3) 120.3(3) 120.3(3) 120.0(3) 120.9'(3) 118.8(3) 120.4(3) 120.8(3) 119.0(3) 120.7(3) 120.2(3) 121.1 (4) 119.9(4) 120.3(4) 120.9(3) 118. 7 (3) 120,2(3) 120.6(3) 11S.6(2)

229

Ni (1) Cl (1) N (1) C(l) Cl(2) C(2) N (2) C(3) C(4) C(5) C (6) C(7) C (8) C (9) C(10) C(ll) C(12) C(13) C(14) C(15) C(16) C(17) C(18) C(19)

. C (20) C (21)

Table 4: Anisotropic displacement parameters (A2 x 103) for nickel complex

143. The anisotropic displacement factor exponent takes the form: -2 pj2 [h2 a*2 U11 + ... + 2 h k a* b* U12]

U11 U22 U33 U23 U13

36 (1) 31 (1) 56(1) 4 (1) 13 (1) 62(1) 50 (1) 92(1) -23 (1) 25 (1) 32 (1) 38 (1) 47(2) -2 (1) 7 (1) 36(2) 46(2) 55 (2) -7 (2) 4 (1) 51(1) 55(1) 62 (1) 22 (1) 14(1) 34(2) 72 (3) 47(2) -4 (2) 6 (2) 32(1) 35(1) 33 (1) 3 (1) 2 (1) 30(2) 82 (3) 38(2) - 8 (2) -1 (1) 43(2) 52 (2) 42 (2) - 8 (2) - 5 (1) 37(2) 44(2) 39(2) -2 (1) 2 (1) 51(2) 34(2) 61(2) - 2 (2) 12(2) 42(2) 36(2) 47(2) 10(1) 5 (2) 37(2) 39(2) 29(2) 2 (1) 2 (1) 33 (2) 36(2) 32 (2) 4 (1) -2 (1) 42(2) 38(2) 44(2) 0(1) 1 (1) 58(2) 32 (2) 62 (2) 2 (2) -7(2) 51(2) 50(2) 60(2) 11(2) 9(2) 45(2) 52 (2) 45(2) 3 (2) 10 (2) 35(2) 40(2) 34(2) 5 (1) 2 (1) 43 (2) 41(2) 29 (2) 5 (1) 5 (1) 45(2) 50(2) 48(2) 2 (2) 10(2) 67(3) 46(2) 53 (2) - 3 (2) 18(2) 80 (3) 39(2) 45(2) - 3 (2) 17(2) 58(2) 45(2) 37(2) o (1) ~ (2) 49(2) 40(2) 25(2) 5 (1) 6 (1) 38(2) 42(2) 32(2) 4 (1) -1(1) .I

230

U12

3 (1) -16(1)

o (1) - 2 (2) 12 (1) - 8 (2) o (1)

11(2) 16(2)

9 (1) 5 (2) 4 (1)

-3 (1) - 2 (1)

1 (1) - 8 (2)

-14(2) - 8 (2) -2 (1) -1 (1)

1 (2) 10(2) -1(2)

-10(2) - 2 (1) - 6 (1)

231

Table 5: Hydrogen coordinates ( x 104) and isotropic displacement parameters

(A2 x 103) for nickel complex 143.

x y z U (eq)

H (1) 5067(34) 1716(21) 1556(22) 38(8) H (2) 7228(40) 1250(22) 840(24) 46 (10) H (3) 7615(48) -155(25) 492(29) 72(12) H(4) 5757(38) -1088(21) 961(24) 46(10) H (6B) 2299 (41) -781(21) 1220(26) 49(10) H (6A) 3475(41) -1345(25) 1782(26) 57(11) H(7A) 1880(38) -1144 (21) . 2941(24) 43 (9) H (7A) 3203(39) -542(22) 3250(25) 49 (9) H (8B) 92 (31) -405(18) 1625(22) 29 (7) H (8A) -734(35) 327(20) 2108(21) 36 (8) H (10) -834(36) -1791(20) 1828(24) 43(9) H (11) -2682(38) -2642(22) 2524(25) 47(9) H(12) -4129(41) -2184(23) 3753(27) 58(11) H(13) -3802(41) -847(21) 4340(26) 52 (10) H (16) -4145(42) 616(23) 3961(25) 54(10) H (17) -3742(48) 1943(26) 4670(30) 71(13) H (18) -:1.465(43) 2458(26) 4976(27) 62(11) H (19) 639(42) 1756(24) 4493(27) 60(11) H (21A) :1.849(36) 465(20) 4107(22) 41 (8) H (21B) 859(34) -362(20) 4102(22) 38(8)


Recommended